fgene-11-623736 January 18, 2021 Time: 16:34 # 1
REVIEW
published: 21 January 2021
doi: 10.3389/fgene.2020.623736
Technological Innovations for
Improving Cassava Production in
Sub-Saharan Africa
Edwige Gaby Nkouaya Mbanjo1, Ismail Yusuf Rabbi1, Morag Elizabeth Ferguson2,
Siraj Ismail Kayondo1, Ng Hwa Eng3, Leena Tripathi2, Peter Kulakow1 and
Chiedozie Egesi1,4,5*
1 International Institute of Tropical Agriculture, Ibadan, Nigeria, 2 International Institute of Tropical Agriculture, Nairobi, Kenya,
3 CGIAR Excellence in Breeding Platform, El Batan, Mexico, 4 National Root Crops Research Institute, Umudike, Nigeria,
5 Department of Global Development, College of Agriculture and Life Sciences, Cornell University, Ithaca, NY, United States
Cassava is crucial for food security of millions of people in sub-Saharan Africa. The
crop has great potential to contribute to African development and is increasing its
income-earning potential for small-scale farmers and related value chains on the
Edited by: continent. Therefore, it is critical to increase cassava production, as well as its
Deepmala Sehgal,
International Maize and Wheat quality attributes. Technological innovations offer great potential to drive this envisioned
Improvement Center, Mexico change. This paper highlights genomic tools and resources available in cassava. The
Reviewed by: paper also provides a glimpse of how these resources have been used to screen
Zhi Zou,
Institute of Tropical Bioscience and understand the pattern of cassava genetic diversity on the continent. Here, we
and Biotechnology, Chinese Academy reviewed the approaches currently used for phenotyping cassava traits, highlighting
of Tropical Agricultural Sciences, the methodologies used to link genotypic and phenotypic information, dissect the
China
Jose Sebastian, genetics architecture of key cassava traits, and identify quantitative trait loci/markers
Indian Institute of Science Education significantly associated with those traits. Additionally, we examined how knowledge
and Research Berhampur (IISER),
India acquired is utilized to contribute to crop improvement. We explored major approaches
*Correspondence: applied in the field of molecular breeding for cassava, their promises, and limitations.
Chiedozie Egesi We also examined the role of national agricultural research systems as key partners for
C.Egesi@cgiar.org sustainable cassava production.
Specialty section: Keywords: cassava, innovations, QTL mapping, association mapping, marker-assisted selection, genomic
This article was submitted to selection, genome-editing
Plant Genomics,
a section of the journal
Frontiers in Genetics INTRODUCTION
Received: 30 October 2020
Accepted: 23 December 2020 The agricultural sector is key to economic growth in Africa. The recent report on the Global
Published: 21 January 2021 Hunger Index indicates that over half of the world’s food-insecure people live in Africa (FSIN,
Citation: 2019). Sustainable agricultural production is imperative to curb food insecurity, reduce poverty,
Mbanjo EGN, Rabbi IY, and impact the livelihood of smallholder farmers (Ojijo et al., 2016; Donkor et al., 2017). Cassava
Ferguson ME, Kayondo SI, Eng NH, is among the six commodities defined by the African Heads of States as strategic crops for the
Tripathi L, Kulakow P and Egesi C continent, given its significant contribution to the livelihoods of African farmers and its potential
(2021) Technological Innovations for for transforming African economies (Feleke et al., 2016).
Improving Cassava Production
in Sub-Saharan Africa. Cassava (Manihot esculenta Crantz) is a root crop grown throughout the tropics by more than
Front. Genet. 11:623736. 800 million people (Nassar and Ortiz, 2010). It can grow with minimal inputs under marginal
doi: 10.3389/fgene.2020.623736 soil conditions and in regions prone to drought. Though mainly cultivated for its starchy roots,
Frontiers in Genetics | www.frontiersin.org 1 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 2
Mbanjo et al. Innovations for Improving Cassava Production
nutrient-dense cassava leaves are also consumed as vegetables in breeding cycle (12 months). New tools and technologies have
many regions of Africa (Spencer and Ezedinma, 2017). Due to the potential to improve the efficiency of conventional breeding,
its long harvest window, cassava roots are used as a food reserve especially when several traits are being selected at the same time.
during periods of food shortage or during the lean season before Modernization of breeding programs, through the application of
harvest of other crops. Although its cultivation has traditionally innovative tools, is vital for more efficient agriculture, especially
been associated with subsistence farming, the crop is gradually in the context of climate change, shrinking resources, land
becoming an industrial crop, which is processed into different scarcity, and increased food demand. Biotechnology and new
products, including bread, pasta, and couscous-like products genomic approaches have the potential to enhance genetic gain,
(Bechoff et al., 2018; Mtunguja et al., 2019). Apart from the food speed up the development of better cultivars, and impact the
industry, cassava starch is used for textiles, the paper industry, livelihoods of smallholder farmers.
in the manufacture of plywood and veneer adhesives, glucose We highlighted the genomic resources available in cassava
and dextrin syrups (Tonukari et al., 2015; Spencer and Ezedinma, and their potential applications. We also reviewed the state of
2017; Waisundara, 2018). knowledge of the phenotyping technologies currently available
Sub-Saharan Africa accounts for 61.1% of the world’s cassava and their shortcomings. Additionally, we explored the potential
production (FAOSTAT, 2020). Although some increase have and implications for integration of technological innovations
been achieved recently, mainly due to expansion in crop area, in cassava genetics and breeding. With farmers being the
cassava average productivity (9.1 t/ha) on the continent is well ultimate beneficiaries, we examined how to ensure regular,
below the average cassava fresh root yield (21.5 t/ha) recorded in sustainable high level of cassava production in sub-Saharan
Asia (Spencer and Ezedinma, 2017; FAOSTAT, 2020). In order Africa; thereby, contributing to food security challenges and
to satisfy the forecasted increase in demand for cassava food improved livelihoods through income generation.
and non-food products, and to harness the enormous potential
offered by the crop, cassava production in sub-Saharan Africa
must be increased (Khandare and Choomsook, 2019; Otekunrin GENOMIC RESOURCES AND THEIR USE
and Sawicka, 2019; FAOSTAT, 2020). Some of the constraints
inherent to its production include pests (Kalyebi et al., 2018; Genomic resources for cassava have increased substantially
Koros et al., 2018) and diseases caused by bacteria (Fanou et al., in recent years. Thousands of simple sequence repeats (SSR)
2018) and viruses (Patil et al., 2015; Alicai et al., 2019) leading markers have been developed from expressed sequence tags
to significant yield losses. Another drawback is the cyanogenic and enriched genomic DNA libraries (reviewed in Ferguson
glucosides, which upon hydrolysis, produces toxic hydrogen et al., 2011). The cassava chromosome-scale reference genome
cyanide (Akinpelu et al., 2011). Cassava is an energy-dense and the re-sequencing of diverse accessions, has identified a
food, mainly composed of starch with low levels of protein and large number of sequence polymorphisms. Single nucleotide
other nutrients important for a balanced diet. Populations that polymorphisms (SNPs), methylation polymorphism, Insertions-
consume cassava as a staple are at high risk of protein, vitamin Deletions (Indels), and structural variants have been identified
A, zinc, and/or iron deficiency (Gegios et al., 2010; Stephenson (Sakurai et al., 2013; Wang et al., 2014; Xia et al., 2014; Bredeson
et al., 2010). These challenges guide the breeding objectives: et al., 2016). A cassava haplotype map harboring 25.9 million
(a) high yield in terms of dry matter per unit land area; (b) SNPs and 19 million indels has been developed (Ramu et al.,
resistance to diseases, such as cassava mosaic disease (CMD), 2017). High-throughput genotyping platforms (e.g., GoldenGate
cassava brown streak disease (CBSD), and cassava bacteria blight assay) have allowed researchers to simultaneously interrogate
(CBB), and pests such as cassava green mites (CGM) and tens of thousands of SNPs at a reduced cost (Ferguson et al.,
whiteflies; (c) improved starch quality and quantity; (d) low 2012). The development of low cost genotyping technologies
hydrogen cyanide potential; (e) improved nutritional value or based on multiplex sequencing platforms, such as genotyping
biofortification; (f) adaption to a wide range of environments; by sequencing, has enabled rapid and accurate high-density
(g) improved plant type for mechanization; and (h) end user fingerprinting using SNP markers (Elshire et al., 2011; ICGMC,
characteristics, including processing, cooking, and organoleptic 2015; Rabbi et al., 2015; Kamanda et al., 2020)5). These resources
properties (Teeken et al., 2020). provide valuable tools that have contributed to genetic research
Conventional breeding has been efficient in providing a and molecular breeding of cassava.
continuous supply of improved cultivars that have resulted in a
dramatic increase in yield of most major crops (Prohens, 2011). Genetic Diversity
Conventional cassava breeding is based on phenotype-based Genetic diversity is of paramount importance in crop
recurrent selection, which relies on the production of full-sib improvement through breeding. The extent and nature of
and/or half-sib progenies followed by successive clonal selection genetic variation existing within cassava landraces and cultivars
stages, including single row trials, preliminary, advanced, and from selected African countries have been assessed using
uniform yield trials (Ceballos et al., 2016). Many cassava varieties molecular markers (Kawuki et al., 2009; Kamanda et al.,
have been developed and released through conventional breeding 2020). These studies have revealed genetic differentiation
(Malik et al., 2020). Breeding cassava is a challenging task due to between African and South American germplasm, as well as
the heterozygous genetic make-up of the crop. The development differentiation between cassava landraces within Africa. Those
of improved varieties is time consuming due to its long from East, South, and Central Africa are somewhat differentiated
Frontiers in Genetics | www.frontiersin.org 2 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 3
Mbanjo et al. Innovations for Improving Cassava Production
from those from West Africa (Kawuki et al., 2009; Ferguson Afonso et al., 2017). A challenge has been to efficiently
et al., 2019; Adjebeng-Danquah et al., 2020). The narrow distinguish the subtle differences within each color group,
genetic base of African cassava breeding lines is attributable to as this is difficult by eye. To overcome this limitation,
intense selection pressure for CMD resistance with recurrent quantification of total carotenoids content by ultraviolet-
selection using few parents and the clonal nature of the crop visible spectrophotometry, as well as identification and
(Turyagyenda et al., 2012; Wolfe et al., 2016; Ferguson et al., quantification of β-carotene and its isomers by high-
2019). Slightly higher genetic diversity has been reported in performance liquid chromatography (HPLC), have been
landraces in comparison to elite accessions (Ferguson et al., employed (Carvalho et al., 2012; Belalcazar et al., 2016).
2019), which is typical of most crops. It is likely that the These approaches are accurate, but have many drawbacks,
genepool of pre-breeding germplasm will become more diverse including cost, time-needed for analysis, labor-intensive
as breeders begin to incorporate variation that responds better to methods, and requirements for laboratory infrastructure
consumer preferences. and trained technical staff, which is not always available to
breeding programs in Africa (Udoh et al., 2017). Alternative
Genetic Redundancy and Breeding portable devices, such as ICheckTM carotene, have been
Impact proven useful for rapid field evaluation and could be
Fingerprinting of cassava accessions using molecular markers valuable in remote areas with no laboratory facilities or
has many applications. Genetic redundancy represents a electricity (Esuma et al., 2016; Jaramillo et al., 2018). Color
challenge to efficiently manage and optimize the conservation instruments designed to quantify the Commission International
of genetic resources in genebanks and breeders’ collections. de l’Eclairage (CIELAB) color parameters have also been
SNP markers have been used to confirm that particular successfully used to evaluate carotenoids content in cassava
accessions are not identical, and others are possible duplicates root samples in an efficient way (Afonso et al., 2017). Near-
(Ferguson et al., 2012). They have also been used to assess infrared spectroscopy (NIRS) is another promising approach
the adoption of improved varieties (Turyagyenda et al., 2012; that has been explored for carotenoids quantification with
Rabbi et al., 2015; Wossen et al., 2017). Molecular markers demonstrated high prediction accuracy (Sánchez et al., 2014;
have an important role to play as farmers frequently give Ikeogu et al., 2019).
different names to the same cultivar or landrace. Thus,
they become difficult to identify, particularly as cassava Cyanogenic Potential
varieties are not easy to distinguish morphologically. This Cassava contains naturally occurring, but potentially toxic
enables the correct assessment of adoption rates, which compounds called cyanogenic glycosides, which release hydrogen
in turn, influences breeding priorities and agricultural cyanide (HCN) on hydrolysis. This can be highly toxic to
policies (Kretzschmar et al., 2018). Molecular markers have humans and animals if not removed through processing. It
also been used to assess the integrity of putative mapping is important that the levels of cyanogenic glycosides are
populations, select true-cross progeny, validate crosses, and measured. Several approaches have been used to quantify
guide parental selection (Rabbi et al., 2012; ICGMC, 2015; cyanide potential, including the titration method (Moriasi
Masumba et al., 2017). et al., 2017; Iliya and Madumelu, 2019), the alkaline picrate
method (Fukushima et al., 2016; Moriasi et al., 2017), and
PHENOTYPING OF KEY TRAITS the metal-based chemosensors (Tivana et al., 2014). Theseapproaches involved multi-step reactions and necessitate trained
Phenotyping is important to support crop breeding. The personnel. The picrate method, although easy to use, is very
acquisition of phenotype data remains a bottleneck hindering slow (minimum 12 h) and the chemicals used are hazardous.
cassava genetic studies and full-deployment of genomics- Recently, it was shown that NIRS could efficiently be used
assisted breeding. Several approaches have been used to to distinguish roots with high or low cyanogenic potential
phenotype breeding lines and germplasm collections for (Sánchez et al., 2014).
nutrition (carotenoids, cyanogenic potential), yield and yield
components (dry matter content), quality (starch physiochemical Dry Matter Content
and functional properties, texture, and pasting properties), biotic Storage root dry matter content (DMC) reflects the proportion
stresses (disease resistance), and root system architecture. In the of useable fresh root yield. DMC is commonly measured using
following section, we detail the current phenotyping strategies either specific gravity through suspension of a root sample in
used in cassava. water and air, or the oven-drying method, which has been the
most widely used method. This is where a representative root
Carotenoids sample is weighted wet and then oven dried to constant weight
Breeding cassava roots with enhanced levels of provitamin A (Fukuda et al., 2010; Teye et al., 2011). The oven-drying method is
carotenoids is a high priority in some breeding programs. tedious when working with a large number of samples. Likewise,
The color of the cassava storage root parenchyma has it is difficult to implement this approach where the source of
been correlated with the total carotenoids content; thus, electricity is unreliable (Teye et al., 2011). Both oven-drying
used to evaluate carotenoids content (Iglesias et al., 1997; and specific gravity could be substituted by NIRS, which has
Frontiers in Genetics | www.frontiersin.org 3 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 4
Mbanjo et al. Innovations for Improving Cassava Production
been shown to predict DMC with a high degree of accuracy springiness, adhesiveness, gel strength, mouldability, elasticity,
(Belalcazar et al., 2016). smoothness, appearance, thickness, and general acceptability
(Rodríguez-Sandoval et al., 2008; Maieves et al., 2012; Rosales-
Starch Physiochemical and Functional soto et al., 2016; An et al., 2019; Ma’Aruf and Abdul, 2020).
Properties Pasting properties of cassava products are key determinants ofquality. Rapid Visco Analyzer (RVA) has been used to evaluate
Physiochemical properties among cassava cultivars determine pasting properties of cassava accessions, and the parameters
root quality attributes important for processing and estimated include peak viscosity, setback viscosity, final viscosity,
consumption. The main constituent of cassava storage roots is pasting temperature, and time to reach peak viscosity. Using
starch, which is composed of amylose and amylopectin. Both RVA, less than five samples can be processed in 1 h (Rosales-
of these play a crucial role in retrogradation, gelling, pasting, soto et al., 2016; Chatpapamon et al., 2019). Although not yet
crystallinity, gelatinization temperature, viscosity, texture, reported for cassava, NIRS has been shown to adequately predict
cooking, eating, and processing quality of cassava (Ayetigbo texture and pasting properties of rice (Meullenet et al., 2002;
et al., 2018). Amylose content (AC) is the most important factor Chueamchaitrakun et al., 2011) and sweet potato (Lu et al.,
influencing cooking and textural quality. The AC in cassava 2006a,b). Therefore, the ability of NIRS to predict cassava pasting
has been estimated using iodine colorimetry (Sandoval-Aldana properties should be explored.
and Fernandez, 2013; Boonpo and Kungwankunakorn, 2017) or
Megazyme amylose/amylopectin assay kit (Chisenga et al., 2019). Biotic Constraints
Iodine colorimetry is prone to inter-laboratory variability due to Different purposes required different discrimination methods for
the complexity of the procedure and relies on the development quantification of disease incidence and/or severity. For example,
of a suitable curve of known amylose to-amylopectin ratios. for breeding purposes, a visual scale of 1–5 from disease-free
The swelling power of starch determines its specific functional to highly diseased may be sufficient, where breeders will only
properties when utilized in food products (Noranizan et al., retain those cultivars with a score lower than two. However,
2010). Swelling power and solubility patterns of cassava flour this is subjective and dependent upon personal perceptions.
have been determined using the Leach (1959) and Kainuma On the other hand, QTL mapping or other applications may
et al. (1967) method, respectively (Chisenga et al., 2019; Ma’Aruf require greater resolution or more accuracy (Garcia-Oliveira
and Abdul, 2020). Starch gelatinization properties (onset, peak et al., 2020). Accurate evaluation and novel approaches for
and conclusion gelatinization temperature, and enthalpy) and cassava disease detection are needed to efficiently assess disease
retrogradation have been determined using differential scanning severity (Chiang et al., 2016). Root necrosis indexes for CBSD
calorimetry (DSC). DSC can be run at a rate of four samples evaluation, which account for the root sample size or their
per hour (Thirathumthavorn and Trisuth, 2008; Tappiban et al., economic value, have been proposed to replace traditional based
2020). Crystallinity measurement requires the use of an x-ray evaluation (Kawuki et al., 2019). Image analysis has similarly been
diffractometer, a piece of complex and expensive equipment used (Garcia-Oliveira et al., 2020; Nakatumba-Nabende et al.,
(Chatpapamon et al., 2019). In sweet potato, it was shown that 2020). Analysis of field images combined with various algorithms,
NIRS could predict most physiochemical and thermal properties including K mean clustering algorithms (Anderson et al., 2015),
of starch with acceptable precision (Lu et al., 2006a,b). NIRS was artificial neutral network (Abdullakasim et al., 2011), and more
sufficiently accurate for the determination of total starch and recently, machine learning techniques and convolutional neutral
amylose in barley in the study of Ping et al. (2013). Meanwhile, network (Owomugisha and Mwebaze, 2016; Sambasivam and
Cozzolino et al. (2013) demonstrated that swelling properties Opiyo, 2020) have provided a more accurate and objective
and water solubility could be determined in whole grain barley assessment of disease severity and incidence. The smart phone-
using NIRS spectroscopy. NIRS technology could potentially based diagnostic system (NURU-AI) is being developed to
be used to predict functional and physiochemical properties of support remote diagnosis by smallholder farmers in Africa for
cassava or cassava-based products. real-time prediction of the state of cassava health (Owomugisha
and Mwebaze, 2016; Ramcharan et al., 2017, 2019).
Texture and Pasting Properties
Texture is a critical factor for consumer acceptance of cassava. Root System Architecture
Sensory analysis has been used for characterizing cassava Uniformity in size and shape of cassava roots is an important
and cassava-based product texture properties. The sensory breeding objective. Routine assessment of storage root size and
descriptors assessed included texture, appearance, odor, taste, shape have relied on visual scores that are both subjective, time
masticability (Akely et al., 2016; Adinsi et al., 2019). The cost consuming, labor-intensive and destructive. Implementation of
associated with training and maintaining a descriptive panel non-invasive approaches include image analysis of roots to
combined with the low throughput of sensory evaluations provide unbiased quantitative data on important root traits. High
has prompted the development of less costly and less time- throughput three-dimensional imaging and non-destructive
consuming approaches. Instrumental methods using a texture methods such as ground penetrating radar (GPR) have recently
analyzer that mimics mastication have been used to evaluate been developed and used for quantifying cassava storage roots
and/or predict the texture of raw, cooked, and processed and their parameters for estimation of size and shape under field
cassava products. The parameters measured include hardness, condition (Delgado et al., 2017, 2019). Kengkanna et al. (2019)
Frontiers in Genetics | www.frontiersin.org 4 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 5
Mbanjo et al. Innovations for Improving Cassava Production
developed cassava shovelomics to evaluate root architecture (based on Manihot esculenta v5 genome assembly corresponding
traits. More recently, Yonis et al. (2020), demonstrated the to chromosome 12 on v5.1 and v6.1) that account for 30 to
feasibility of root phenotyping using image capture and analysis. 60% of variation in genetic resistance. Additional regions with
Routine implementation of these new phenotyping solutions small effects, including one on chromosome 9 that co-located
offers new opportunities for cassava breeders to efficiently and with the CMD1 resistance were also reported (Wolfe et al., 2016;
precisely select and release cultivars with root architecture that Table 3). The study of Wolfe et al. (2016) substantiated bi-
are favorable for harvesting, processing, as well as selection parental mapping studies (Akano et al., 2002; Okogbenin et al.,
for early bulking characteristics. The development of robust 2012) reporting the single major gene, CMD2, determining CMD
phenotyping technologies for large-scale phenotyping with resistance and a second QTL, CMD3, closely linked to CMD2.
increased precision at reduced cost will enable efficient screening A key outcome of the Wolfe et al. (2016) study was the lack of
of larger populations. Precise phenotyping is invaluable for other major-effect loci. Likewise, significant interactions between
carrying out downstream analysis, including characterization of the significant SNPs were disclosed. Nzuki et al. (2017) found
genetic factors that contribute to phenotypic variation. two QTLs associated with CMD on chromosomes 12 and 14.
The percentage of variation explained (PVE) by these QTLs were
13.01 and 13.36%, respectively. The QTL on chromosome 12
CONNECTING GENOTYPIC AND was confirmed as the QTL linked to the CMD2 locus. Masumba
PHENOTYPIC INFORMATION AND et al. (2017) detected two highly significant CMD resistance
QTLs on chromosome 12 defining the CMD2 locus, as well
DISCOVERY OF USEFUL GENES as additional putative QTLs on other chromosomes. Similar
AND/OR QUANTITATIVE TRAIT LOCI observations were made by Garcia-Oliveira et al. (2020) who
found two closely linked loci on chromosome 12 and additional
Genotype variations have been linked to corresponding QTLs on chromosomes 9 and 10. Further dissection of the major
differences in phenotypes and the genetic basis of the phenotypic QTL on chromosome 12 (Manihot esculenta v5.1 and 6.1 genome
variability evaluated to identify genes and/or quantitative trait assembly) revealed the presence of two possible epistatic loci
loci (QTLs) associated with the traits of interest. Two approaches and/or multiple resistance alleles, which may account for the
have been used for genotype-phenotype association: (1) classical difference between moderate and strong disease resistance in the
QTL mapping using experimental populations derived from germplasm (Masumba et al., 2017; Nzuki et al., 2017; Garcia-
bi-parental crosses with contrasting phenotypes (Collard et al., Oliveira et al., 2020; Table 2). Gaps in the pseudochromosome
2005); and (2) genome-wide association studies (GWAS) 12 region containing the CMD2 loci caused by highly repetitive
mapping that use germplasm collections and incorporate DNA might explain the two separate loci (Kuon et al., 2019).
historical events that have occurred during domestication of Somo et al. (2020) identified QTLs associated with CMD severity
the crop. Classical QTL analysis in cassava has been made across all chromosomes, except on chromosomes 4, 5, 6, 8, 11, 13,
possible with the development of numerous genetic linkage maps and 18. They validated the presence of the previously reported
(Table 1) and use of statistical approaches that have resulted CMD2 locus and detected another major QTL on chromosome
in several QTLs identified for economically important traits 16. Nzuki et al. (2017) and Rabbi et al. (2020) confirmed the role
(Table 2). With the decreasing cost of DNA sequencing, new of CMD2 loci as the major gene for CMD resistance and reported
genotyping technologies and their improved accuracy, GWAS two additional loci on chromosome 14 (Table 3).
is being increasingly used for genetic analysis of traits. GWAS
has enabled the exploration of allelic diversity that exists in Cassava Brown Streak Disease
natural populations, the discovery of beneficial alleles in several Quantitative trait loci associated with CBSD root necrosis were
crops. Marker-trait associations in cassava have focused on a few identified on chromosomes 5, 11, 12, and 15 by Nzuki et al.
important traits of interest. (2017). The detected QTLs explained up to 10.18% of the PVE.
Masumba et al. (2017) reported two consistent QTLs linked to
Cassava Mosaic Disease resistance to CBSD-induced root necrosis on chromosomes 2
The dominant gene CMD2 underlying CMD resistance was and 11, as well as a putative QTL on chromosome 18. Further
discovered in the farmer-preferred Nigerian landrace TME 3 additional putative QTLs were detected on other chromosomes
(Akano et al., 2002). A new linked QTL underlying CMD (3, 4, 5, 6, 7, 10, 12, 15, and 16). In addition to QTLs found on
resistance named CMD3 that explained 11% of the phenotypic chromosome 8 and 18, new putative QTL for CBSD root necrosis
variance (PVE) was later identified 36 cM away of CMD2 gene was identified on chromosome 14 (Garcia-Oliveira et al., 2020).
by Okogbenin et al. (2012). The qualitative nature of CMD Seven QTLs associated with CBSD foliar symptoms were found
resistance was confirmed in subsequent studies, and a single locus on chromosomes 4, 6, 15, 17, and 18. The most significant QTL
with a large effect in the vicinity of the previously mapped CMD2 explained 8.45% of the PVE (Nzuki et al., 2017). The study of
locus was uncovered (Rabbi et al., 2014; Echefu et al., 2016; Masumba et al. (2017) revealed several QTLs on all chromosomes
Table 2). The first GWAS mapping study in cassava conducted linked with CBSD foliar symptoms, the most interesting of
by Wolfe et al. (2016) identified 198 significant SNPs associated which was found on chromosome 2 and explained 4.6% of the
with CMD severity on 14 chromosomes. The significant SNPs PVE. Garcia-Oliveira et al. (2020) identified one major QTL
were mostly concentrated in a single region on chromosome 8 on chromosome 18 that explained 12.87% of the phenotypic
Frontiers in Genetics | www.frontiersin.org 5 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 6
Mbanjo et al. Innovations for Improving Cassava Production
TABLE 1 | Summary of the published genetic linkage maps of cassava.
Population Mapped Linkage groups/ Map length (cM) References
markers chromosomes
TMS30572 (♀) × CM2177-2 32 RFLPs, 30 RAPDs, 20 932.6 (female) Fregene et al., 1997
(♂) F1 (n = 90) 3SSRs, 3 Isoenzymes
107 RFLPs, 50 RAPDs, 24 1220 (male)
1 SSR
TMS30572 (♀) × CM2177-2 (♂) 100 SRRs 22 1236.7 Okogbenin et al., 2006
F2 (n = 268)
Huay Bong × Hanatee 510 SSRs 23 1420.3 Sraphet et al., 2011
F1 (n = 100)
Hanate (♀) × Huay Bong (♂) 303 SSRs 27 1328 Whankaew et al., 2011
F1 (n = 100)
Namikonga (♀) × Albert (♂) 434 SNPs, 134 SSRs 19 1837 Rabbi et al., 2012
F1 (n = 130) (one step)
348 SNPs, 128 SSRs 18 1541
(two steps map)
IITA-TMS-4(2)1425 (♂) x 6756 SNPs 19 3771 Rabbi et al., 2014
IITA-TMS-011412 (♂)
F1 (n = 180)
KU50 × SC 124 2331 SNPs, 537 Indels, 20 1970.40 Xia et al., 2014
F1 (n = 85) 164 methylation
markers
Huay Bong (♀) × Hanatee (♂) 2110 SNPs 19 1785.62 Pootakham et al., 2014
F1 (n = 100)
TMS30572 (♀) × CM2177-2 (♂) 2141 SNPs 18 2571 Soto et al., 2015
F1 (n = 132)
9 bi-parental F1 crosses 22,403 SNPs 18 2412 ICGMC, 2015
1 self-pollinated cross (S1)
n = 1740
Kibora (♀) × AR37-80 (♂) 1974 SNPs 21 1698 Nzuki et al., 2017
F1 (n = 106)
Namikonga (♀) × Albert (♂) 943 SNPs 18 1776.2 Masumba et al., 2017
F1 (n = 240)
AR40-6 (♀) × Albert (♂) 2125 SNPs 18 1730 Garcia-Oliveira et al.,
F1 (n = 130) 2020
SNP, single nucleotide polymorphism; RAPD, random amplified polymorphism DNA; RFLP, restriction fragment length polymorphisms; SSR, simple sequence repeat.
variation, but at a slightly different position. These same authors have been associated with either root necrosis or foliar symptoms,
also reported a minor QTL on chromosome 11 (Table 2). The supporting the notion that resistance to foliar and root symptoms
GWAS approach was used by Kayondo et al. (2018) to unravel the of CBSD are largely under different genetic control (Masumba
genetic architecture of CBSD. The polygenic control mechanism et al., 2017; Garcia-Oliveira et al., 2020). The detected QTLs were
of resistance for CBSD and its instability across the environment not consistent across studies. Different mapping populations
was highlighted. Eighty-three (83) loci associated with foliar were used in the case of experimental populations. There might
symptoms at 3 months after planting (MAP) were identified on be some variations in CBSD response. Furthermore, QTLs tend
chromosome 11. The top SNPs explained 6% of the phenotypic to be population specific in bi-parental populations. Likewise,
variance. Significant SNPs were identified on chromosomes 11, different environmental conditions, population size (106–1986
4, and 12 for foliar severity score at 6 MAP. Recently, Somo samples), and the subjectivity of the scoring system used for data
et al. (2020) using a diverse panel of breeding lines, identified collection could have affected QTL detection and localization.
QTLs conferring resistance to CBSD on chromosomes 9 and 11 The use of different versions of the cassava genome assembly
that accounted for 9 and 5% of PVE, respectively. Nine markers makes it challenging to compare some of these results. Therefore,
representing four loci on chromosomes 2, 3, 8, and 10 associated these QTLs should be mapped onto the most recent version of the
with resistance to CBSD for root necrosis were also reported cassava reference genome.
by these authors (Table 2). Putative regions on chromosomes
11 and 15 were shown to be associated with both CBSD foliar Cassava Green Mite and Related Traits
and root necrosis symptoms (Nzuki et al., 2017), suggesting that Nzuki et al. (2017) detected on chromosomes 5 and 10 QTLs
CBSD root necrosis and CBSD foliar symptoms are most likely associated with cassava green mite (CGM) with maximum PVE
to be influenced to some extent by the same gene(s) or by closely of 10.56 and 10.08%, respectively. Recently, 95 SNP markers
linked genes at this locus. However, most of the QTLs reported significantly associated with CGM resistance were reported by
Frontiers in Genetics | www.frontiersin.org 6 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 7
Mbanjo et al. Innovations for Improving Cassava Production
TABLE 2 | Summary of the QTL studies of cassava using controlled populations.
Traits Mapping Population Markers Used Key Findings References
CBB TMS 30572 (♀) × CM 68 RFLPs (female map) Eight (8) QTLs identified on five LGs in the female map Jorge et al., 2000
2177-2. (♂) 15 RFLPs (male map) explaining 9 - 20%. In the male map, four QTLs
F1 (n = 150) explaining 10.7–27.1% of the PVE.
CBB TMS 30572 (♀) × CM Female map (95 Eight (8) QTLs detected. Alleles coming from both Jorge et al., 2001
2177-2 (♂) RFLPs, 36 RAPD, 3 parents contribute to the resistance in the progenies.
F1 (n = 150) isoenzymes, 3 SSRs, 3
EST, 2 unknown genes)
Male map (89 RFLPs,
41 RAPDs, 4 unknown
genes, 1 EST)
CMD TME 3 × TMS 30555 186 SSRs First report of the CMD2 locus associated with CMD Akano et al., 2002
F1 (n = 80) resistance and identification of SSRs linked to the CMD
resistance genes CMD2.
DMC MCOL 1684 × THAI 1 98 SSRs Six (6) QTLs on four different linkage groups explaining Kizito et al., 2007
F2 (n = 199) 14 to 40% of the PVE.
Cyanogenic Two (2) QTLs on two different LGs explaining 7% and
glycoside (CN) 20% of the PVE, respectively.
Anthracnose TME 117 (♀) × TMS 53 RAPDs Using bulk segregant analysis, two markers (OPAF2 Akinbo et al., 2007
92/0326 (♂) and OPF06) flanking the CAD1 genes were identified
F1 (n = 60) that explained 33.7 and 27.4% of the phenotypic
variation.
Fresh foliage TMS 30572 (♀) × CM 122 SSRs Three (3) QTLs for fresh foliar detected on three linkage Okogbenin et al., 2008
2177-2 (♂) groups. The PVE ranged from 5.5 to 31.1%.
F2 (n = 268)
Harvest index Three (3) QTLs associated with harvest index on three
LGs explaining 2.10–6.67% of the phenotypic variation.
QTL with major effect accounted for 36.9%.
Dry yield root Three (3) QTLs for dry root yield on three LGs (1, 3, 13)
accounting for 6.0–9.2% of the PVE.
Cyanogenic Hanatee (♀) × Huay 303 SSRs Five (5) QTLs mapped on LG2, 5, 10, 11 explained 16.1 Whankaew et al., 2011
potential Bong (♂) to 26% of the PVE. The most significant association
F1 (n = 100) was on LG10.
CGM Four backcross 500SSRs Using bulk segregant analysis, three markers (SSRY11, Choperena et al., 2012
populations (CW 65, SSRY 346, and NS1099) associated with CGM
66, 67, 68) resistance were detected.
Starch pasting Hanatee (♀) × Huay 510 SSRs Fifteen (15) QTLs affecting five starch pasting viscosities Thanyasiriwat et al.,
viscosity Bong (♂) identified when average values of the three 2013
F1 (n = 100) environments were used. The PVE ranged from 10.0 to
48.4%. Data from separate environment revealed 48
QTLs for seven starch pasting properties, explaining 6.6
to 43.7% of the phenotypic variation.
Starch pasting Huag Bong 2110 SNPs Single co-localize QTL for pasting temperature and Pootakham et al., 2014
properties (♀) × Hanate (♂) pasting time detected on LG7. The PVE explained was
F1 (n = 100) 44.7 and 24.3% for pasting temperature and pasting
time, respectively. Additional QTL for starch pasting
time found on LG10 (PVE = 22.5%).
CMD IITA = TMS-4(2) 1425 6756 SNPs Single SNP underlying CMD resistance in the vicinity of Rabbi et al., 2014
(♂) x the previously mapped CMD2 locus that explain 74% of
IITA-TMS-011412 (♂) the variance
F1 (n = 108)
CMD TMS961089A× TMEB117 4394 SNPs Single SNP underlying CMD resistance in the vicinity of Echefu et al., 2016
the previously mapped CMD2 locus that explain 74% of
the variance
Fresh starch Huag Bong 510 SSRs Nine (9) QTLs on seven LGs, including 6, 7, 9, 11, 13, Sraphet et al., 2017
content (♀) × Hanate (♂) and 11 that explained 11.3 to 27.3% of the phenotypic
F1 (n = 100) variation. Six of the QTLs were location specific while
two were found across environments.
CBSD root necrosis Namikonga (♀) × Albert 943 SNPs Two QTLs consistent across seasons/sites on Chr11 Masumba et al., 2017
(♂) and 2 and a putative QTL on Chr18. Further additional
F1 (n = 240) QTLs on Chr3, 4, 5, 6, 7, 10, 12, 15, and 16
(Continued)
Frontiers in Genetics | www.frontiersin.org 7 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 8
Mbanjo et al. Innovations for Improving Cassava Production
TABLE 2 | Continued
Traits Mapping Population Markers Used Key Findings References
CBSD foliar Several QTLs revealed on all chromosomes. The most
symptoms interesting QTL was on Chr2 and has a PVE of 4.6%.
CMD Two (2) QTLs identified on Chr12 that explain 16.43 and
13.5% of the phenotypic variation. Additional QTLs
detected on all other chromosomes except Chr11 and
14.
CBB TMS30572 × CM2177- 2571 SNPs Five-strain specific QTLs for resistance to Xam Sedano J. C. S. et al.,
2 explaining between 15.8 and 22.1% of the PVE 2017
F1 (n = 117)
CBSD root necrosis Kibora (♀) × AR37-80 1974 SNPs Three (3) QTL associated with CBSD root necrosis on Nzuki et al., 2017
(♂) Chr5, 11, and 12 explaining 2.62–10.18% of the
F1 (n = 106) phenotypic variation
CBSD foliar A total of seven QTLs detected on Chr4, 6, 17 and 18
symptoms explaining 2.64–8.45% of the PVE
CMD Two (2) QTLs detected on Chr12 and Chr14 with a
maximum PVE of 13.01 and 13.36%, respectively.
Leaf pubescence TMEB778 42204 SNPs Seventy-one (71) SNPs significantly associated with leaf Ezenwaka et al., 2020
(♀) × TMEB419 (♂) pubescence on Chr12. The top significant SNP
F1 (n = 109) explained 26% of the phenotypic variation.
Stay green A total of 126 SNPs associated with stay green. The
PVE ranged from 20 to 30%.
CMD AR40-6 (♀) × Albert (♂) 2125 SNPs Provide evidence for two QTL or multi-allelic variants Garcia-Oliveira et al.,
F1 (n = 130) influencing CMD resistance within the locality of CMD2 2020
locus. Additional putative QTLs on Chr 9 and 10
CGM Five (5) minor effect QTLs specific to the planting stage.
Four QTLs detected at 3 MAP on Chr 5, 9, 13, and 18.
PVE ranging from 1.35 to 11.06%, while one QTL
detected on Chr16 at 6 MAP accounting for 7.27% of
the PVE.
CBSD foliar One minor QTL on Chr11 accounting for 2.5 %
symptoms phenotypic variation. One major QTL identifying on
Chr18 explaining 12.87% of the phenotypic variation
CBSD root necrosis New putative QTL for CBSD root necrosis identified on
Chr14
SNP, single nucleotide polymorphism; RAPD, random amplified polymorphism DNA; RFLP, restriction fragment length polymorphisms; SSR, simple sequence repeat;
EST, expressed sequence tag; CBSD, cassava brown streak disease; PVE, phenotypic variance; LG, linkage group; Chr, Chromosome.
Ezenwaka et al. (2020). The significant markers concentrated in resistance to CGM on chromosomes 8 and 12 (Rabbi et al.,
a single region of the left arm side on chromosome 12. The 2020; Table 3).
variance explained by the significant markers ranged from 18 to
31%. Garcia-Oliveira et al. (2020) detected five QTL for CGM Cassava Bacterial Blight
resistance at 3 and 6 MAP, all with minor effect on chromosomes Two QTLs associated with cassava bacterial blight (CBB) caused
5, 9, 13, and 18. While the QTLs on chromosome 9 was found by Xanthomonas axonopodis pv. manihotis (Xam) were reported
in all four environment tested, the QTL on chromosome 8 co- on linkage group (LG) 4 and LG8 that explained 12.6 and 10.9%
localized with previously reported marker for CGM resistance of the field resistance to CBB, respectively (Sedano J. S. et al.,
(Table 2). The first GWAS to identify SNPs linked to CGM and 2017). Another study conducted by Sedano J. C. S. et al. (2017)
CGM-related traits was performed by Ezenwaka et al. (2018) found five strain-specific QTLs conferring resistance to Xam that
who found 35 significant SNP markers, including 12, 17, 5, explained 15.8 and 22.1% of the phenotypic variance. Three of the
and 1 associated with CGM, leaf pubescence, leaf retention, associated QTLs were found to be effective against Xam318 strain
and stay green, respectively. All the significant markers were and explained 17.3 to 18.8% of the phenotypic variance, while two
found on chromosome 8, except the SNP associated with stay were detected for Xam681 and accounted for 15.8–22.1% of the
green, which was identified on chromosome 13. Some of the phenotypic variance (Table 2).
significant SNP markers on chromosome 8 reported by these
authors were also detected in the recent study of Rabbi et al. Cassava Root Rot
(2020) who identified other putative loci on chromosomes 1, The complex nature of cassava root rot disease (CRR) was
12, and 9. Association analysis of CGM-related traits, apical highlighted by Brito et al. (2017), who identified 38 significant
pubescence identified significant loci on five chromosomes, SNPs associated with CRR. Of these, 8 and 22 were related to the
two of which co-located in the same regions underlying severity of dry root rot in the pulp and peel, respectively, while
Frontiers in Genetics | www.frontiersin.org 8 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 9
Mbanjo et al. Innovations for Improving Cassava Production
TABLE 3 | Summary of the GWAS studies of cassava.
Trait Population/size Markers used Key findings References
CMD African breeding lines 42 113 SNPs A total of 198 SNPs mostly concentrated in a single Wolfe et al., 2016
n = 6128 region on Chr8, explaining 0.55 to 22% of PVE.
Dry matter content Genetic Gain Collection 72 729 SNPs Major locus at 24.1 Mbp of Chr1 Rabbi et al., 2017
n = 672
Provitamin A Major association regions at 24.1 and 30.5 Mbp on
Chr1
Root rot severity Fusraium Cassava germplasm 14 094 SNPs Twenty-two (22) SNPs on Chr2, 5, 6, 8, 9, 10, 12, 17, Brito et al., 2017
(peel) bank (EMBRAPA) 18, and ‘A fragments’ most significantly found on
n = 263 chromosome 10. The PVE ranged from 4.66 to 9.54%.
Root rot severity Eight (8) significant SNPs on Chr2,11.14,15,16. The
Fusarium (Pulp) most significant SNP was on Chr 11. The PVE range
from 4.86 to 6.13%.
Root rot severity One (1) SNP on Chr13 explaining 7.99% of the
Phytophthora (peel) phenotypic variation
Root rot severity One (1) SNP on Chr8 explaining 6.80% of the
Phytophthora (peel) phenotypic variation.
Root rot severity Three (3) SNPs on Chr5, 13, and genomic fragment "B"
Botryosphaeriaceae (peel) explain 8.90–10.09% of the phenotypic variation.
Root rot severity Three (3) SNPs on Chr9 and 12, explaining 5.47–7.04%
Botryosphaeriaceae (pulp) of the phenotypic variation.
Cassava green mite Advanced breeding 61 307 SNPs Twelve (12) significant loci on Chr8. The top significant Ezenwaka et al., 2018
lines explained 7% of the PVE.
n = 845
Leaf pubescence Seventeen (17) loci on Chr8. PVE ranged from 4 to 7%.
Leaf retention Five (5) loci on Chr8 each explaining 4% of the
phenotypic variation.
Stay green One (1) significant SNP on Chr13 explaining 4% of the
phenotypic variation.
Carotenoid content SC9 × white-root SC 104 059 SNPs Eighty-four (84) SNPs and 694 genes on all Luo et al., 2018
F1 (n = 98) chromosomes except Chr5.
Storage root quality Diversity Panel 19 850 SNPs Four (4) association signals, including two SNPs on Zhang et al., 2018
n = 158 Chr5, and 15 for starch content and two SNPs on
Chr14 and 16 for dry mass content.
Yield components Seven (7) association signals. Number of storage root
(3, Chr4, Chr 5, Chr 9); storage root weight (1, Chr 2),
dry mass weight (3, Chr 2).
Morphological Eleven (11) association signals. Stem diameter (6; Chr3,
characteristics Chr6, Chr7, Chr14). First branch height (5; Chr2, Chr3,
Chr9, Chr10).
Leaf characteristics Fourteen (14) association signals. Lobular length (4;
Chr4, Chr5, Chr6, Chr18). Lobular width (3, Chr3, Chr5,
Chr10). Petiole length (1, Chr1), Leaf aspect ratio (6,
Chr1, Chr3, Chr6, Chr10, Chr14).
CBSD foliar severity Diversity panel 63 016 SNPs Eighty-three (83) significant SNPs located on Chr11 the Kayondo et al., 2018
(3 months) n = 1986 top SNPs explained 6% of the observed significant
variation.
CBSD foliar severity The significant SNPs were located on Chr11, Chr4, and
(6 months) Chr12.
All-trans β-carotene GS training population 87 380 SNPs Seven (7) significant SNPs with SNP effect ranging from Ikeogu et al., 2019
n = 594 0.328 to 0.592.
Lutein Twelve (12) significant SNPs and 11 unique loci with
SNP effects ranging from 0.025 to 0.034
Total carotenoids Twenty (20) significant SNPs with SNP effect ranging
content from 0.435 to 0.749.
Violaxanthin One (1) significant SNP with an effect of 0.01
13-cis beta-carotene Twenty-three (23) significant SNPs with SNP effect
ranging from 0.039 to 0.058.
15-cis beta-carotene Twenty-one (21) significant SNPs with SNP effect
ranging from 0.008 to 0.011.
(Continued)
Frontiers in Genetics | www.frontiersin.org 9 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 10
Mbanjo et al. Innovations for Improving Cassava Production
TABLE 3 | Continued
Trait Population/size Markers used Key findings References
9-cis beta-carotene Eighteen (18) significant SNPs with SNP effects ranging
from 0.032 to 0.045.
CMD Diverse breeding lines 84 434 SNPs Genetic association detected on Chr1, 2, 3, 7,9,10, 12, Somo et al., 2020
panels 15, 15, 16, and 17. The presence of CMD2 locus was
n = 432, n = 408, validated. Another major CMD QTL was detected on
n = 402 Chr16.
CBSD root severity Nine (9) SNPs on Chr2, 3, 8, 10 were significant.
CBSDS| CMDS One QTL on Chr4 and 2 on Chr12 significantly
associated with both CBSDS/CMDS resistance.
Waxy starch Diversity panel 20,956 SNPs 10 association signals on Chr2 do Carmo et al., 2020
n = 382
Cassava mosaic disease Elite breeding clones 101,521 SNPs CMD2 locus and other loci on Chr14 were captured. Rabbi et al., 2020
n = 5130 SNP effects ranged from −0.23 to 0.82
Cassava green mite Four (4) regions associated with CMG on Chr1, 8, 12,
19. SNP effect size ranged from −0.18 to 0.10
Apical pubescence Significant association on Chr8, 9,11, 12, 16 with effect
size from −0.19 to 0.08.
Carotenoid content Eight (8) significant associations on Chr1, 5, 8, 15, and
16 with SNP effect between −0.007 and 0.49. Major
locus on Chr1 Around 24.1, 24.6 and 30.5 Mb.
Dry matter content Five (5) significant SNPs on Chr1, 6, 12, 15, 16. SNP
effect between −1.32 and 0.85. The most significant
SNP occurred on Chr1 around 24.64 Mb.
Harvest index Two (2) genomic regions associated with this trait on
Chr2 and Chr12, with SNP effect of −0.04 and −0.03,
respectively.
Morphological traits Eight (8) association signals detected. Outer cortex
color 2 SNPs on Chr1 (3.05 Mbp) and Chr2 (6.56 Mbp).
A single genomic region associated with periderm color
on Chr3 (4.54 Mbp). 2 SNPs association signal for plant
type on Chr1 (2.19 Mbp and 25.30 Mbp). Three Loci
associated with stem color and Chr2 and Chrx 8 which
contain the most significant at around 13.6 Mbp.
Leaf morphology Seven (7) association signals. Leaf petiole color single
genomic region on Chr1 (23.45 Mbp); the same region
was associated with mature leaf greenness. 3
association signals for apical leaf color (Chr2, 3, 8). For
leaf shape two major loci on Chr 15 around 10.27 and
20.57 Mbp.
Cyanogenic glycoside Test population 27,054 SNPS Two significant peaks on chromosome 16 and 14, Ogbonna et al., 2020
(n = 1246) explaining 36 and 8% of the variance, respectively.
Validation population
(n = 636)
CBSD, Cassava Brown Streak Disease severity; CMD, Cassava Mosaic Disease.
the other eight were associated with soft root rot and black root identified 84 SNPs distributed in all chromosomes, except
rot (Table 3). chromosome 5, associated with carotenoid traits. Ikeogu
et al. (2019) identified 42 unique SNPs significantly
Carotenoids and Storage Root Color associated with variation in total carotenoid content on
Three QTLs that control the content of carotenoids and chromosomes 1, 2, 4, 13, 14 and 15. Additional regions
four QTLs linked to the color of cassava roots pulp were for variation in the total carotenoid content, as well as
identified by Morillo et al. (2013) (Table 2). Root yellowness the individual carotenoids, were uncovered. Some regions
resulting from carotenoid accumulation elucidated through associated with more than a single carotenoid were identified,
GWAS has revealed major association regions that govern suggesting the possibility of pleiotropic effects (Table 3).
this trait around 24.1 and 30.5 Mbp of chromosome 1 (Rabbi The level of phenotypic variability, the SNP frequencies
et al., 2017). More recently, using a larger diversity panel, and distributions, and the difference in sample size (98–
Rabbi et al. (2020) confirmed these previous findings. They 5130 samples) between the panels used for GWAS might
found five new genomic regions associated with carotenoid explain the differences in QTLs detected. The wider coverage
content on chromosomes 5, 8, 15, and 16. Luo et al. (2018) of diversity could increase the detection of true novel
Frontiers in Genetics | www.frontiersin.org 10 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 11
Mbanjo et al. Innovations for Improving Cassava Production
associations, while small sample size could lead to some groups (LG4, 6, 7, 9, 11, 13, and 16). Among these, six QTLs
spurious associations. were location-specific, and three QTLs were detected across three
environments. The percentage of phenotypic variance explained
Cyanogenic Glucosides by the QTLs ranged from 11.3 to 27.3% of the phenotypic
Kizito et al. (2007) reported two QTLs (SSRY105, SSRY42) on variation (Sraphet et al., 2017; Table 2). Using GWAS, 10 SNP
two different linkage groups controlling cyanogenic glucosides. associated with waxy phenotypes were identified on chromosome
The two QTLs explained 7% and 20% of phenotypic variation, 2 that co-located in genic regions that included five known
respectively. Also, both QTLs showed additive effects. Five QTLs genes and five genes of unknown function (do Carmo et al.,
associated with cyanogenic potential were identified across four 2020; Table 3).
linkage groups, including LG2, 5, 10, and 11 (Whankaew et al.,
2011). The percentage of phenotypic variance explained from Agronomy and Physiology Traits
all detected QTLs ranged from 15.9 to 26.0% (Table 2). One Eight QTLs associated with fresh root yield were identified on
(SSRY42) of the two QTLs reported by Kizito et al. (2007) seven linkage groups (LG1, 2, 6, 9,12,13, and 16) from a bi-
was found in this latest study. More recently, Ogbonna et al. parental mapping population, of which two QTLs on linkage
(2020) reported two genomic regions, in chromosomes 16 group 16 were found across two environments. These QTLs
and 14, associated with hydrogen cyanide potential in cassava. explained 12.9 to 40% of phenotypic variation (Sraphet et al.,
The most significant marker was found on chromosome 16. 2017; Table 2). Zhang et al. (2018) reported 36 loci related to
Chromosomes 16 and 14 tagged SNPs explained 36 and 8% 11 agronomic traits, including leaf characteristics, morphological
of phenotypic variance, respectively (Table 3). The significant characteristics, yield components, and root quality that were
region on chromosome 14 coincides with previously reported identified by GWAS analyses. They found seven SNPs associated
cyanide associated QTL reported by Kizito et al. (2007). with yield components that explained about 14.95% of the
phenotypic variance on average. Morphological characteristics
Dry Matter Content exhibited 11 association signals and explained 12.23 to 20.86%
Six QTLs detected on four different LGs were reported to of the phenotypic variance. A total of 14 SNPs were identified
control DMC using a bi-parental mapping population. Individual from leaf characteristics, which explained 11.9 to 22.6% of the
QTL explained 14 to 40% of the variance. It was shown that phenotypic variance. Genomic regions associated with harvest
additive, dominance, and overdominance effects play a role index were uncovered on chromosomes 2, 3, 4, 6, 8, 9, 12, 14,
in the expression of this trait (Kizito et al., 2007; Table 2). and 15. Two association signals for outer cortex color were found
Using genome-wide association mapping, DMC was found to be on chromosomes 1 (3.05 Mbp) and 2 (6.56 Mb). A single genomic
associated with a major locus occurring on a 24.1 Mbp region region on chromosome 3 (4.54 Mbp) has been linked to periderm
of chromosome 1 (Rabbi et al., 2017). This locus was confirmed color. Two association signals were found for plant types on
by the recent study of Rabbi et al. (2020), who reported another chromosome 1 (2.19 and 25.30 Mbp). Five loci associated with
major locus on chromosome 6 as well as three additional loci on stem color variation were reported on chromosomes 2 and 8. The
chromosomes 12, 15, and 16 (Table 3). most significant loci were around 13.6 Mbp on chromosome 8.
A single genomic region at around 23.45 Mbp on chromosome 1
Starch and Starch Quality was found concomitantly associated with leaf petiole color and
The study of Thanyasiriwat et al. (2013) revealed the complex mature leaf greenness. Variation in leaf color was found to be
inheritance of starch pasting properties. Using average values of associated with three loci occurring on chromosomes 2, 3, and
three environments, these authors reported 15 QTLs on LG1, 8. At around 10.27 and 20.57 Mbp on chromosome 15, two loci
4, 6, 7, 8, 10 12, 13, 14, 16, and 18 affecting five starch pasting associated were identified (Rabbi et al., 2020; Table 3). Seventy-
viscosities (peak viscosity, hot paste viscosity, cool paste viscosity, one (71) markers significantly associated with leaf pubescence
set back, and pasting temperature). The detected QTLs explained were identified by Ezenwaka et al. (2020) on chromosome 12. The
10.0 to 48.4% of the phenotypic variance. Based on analysis of variance explained by these significant markers ranged from 18
each environment, 48 QTLs significantly associated with seven to 26%. The same study of Ezenwaka et al. (2020) reported 126
starch pasting viscosities (peak viscosity, hot paste viscosity, SNP markers associated with stay green on chromosome 12, and
break down, cool paste viscosity, setback, pasting time, and the variance explained by the detected QTLs ranged from 20 to
pasting temperature) were detected on all LGs except LG2, 4, 30% (Table 2).
and 7. The PVE ranged from 6.6 to 43.7. Thanyasiriwat et al. Numerous favorable alleles, functional loci or regions linked
(2013) also reported two major QTLs on LG1 and LG6 for to traits of interest have been identified through marker-
pasting temperature, which accounted for more than 70% of the trait associations and their phenotypic contribution identified,
phenotypic variation. Pootakham et al. (2014) identified a single giving a glimpse to the genetics underlying phenotype variation.
co-localized QTL controlling pasting temperature and pasting However, the designation of linkage groups/chromosomes are
time on LG7. The major QTL explaining 44.7 and 24.3% of sometimes different, making it challenging to align QTL detected
the phenotypic variance. These authors reported additional QTL in different studies (Garcia-Oliveira et al., 2020). Synchronization
associated with starch pasting time on LG 10 that accounted of cassava QTL information and the development of a cassava
for 22.5% of the phenotypic variance. A total of nine QTLs QTL repository is needed. A cassava QTL database that contains
controlling fresh starch content was identified on seven linkage QTL information systematically aligned to the cassava reference
Frontiers in Genetics | www.frontiersin.org 11 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 12
Mbanjo et al. Innovations for Improving Cassava Production
genome, as well as information about germplasm and genetic DMC, to low-plex-high-throughput marker assay2. The reliability
material used in the QTL studies, will be a useful resource for and utility of these markers are currently being evaluated in
both cassava geneticists and breeders (Ni et al., 2009; Yonemaru different genetic backgrounds and across different environments.
et al., 2010; Said et al., 2015). A benefit envisioned from the A KASP assay has recently been employed to develop and
identification of molecular markers tightly linked to traits of validated diagnostic markers for HCN content (Ogbonna et al.,
interest is their deployment in breeding as an indirect selection 2020). The use of shared genotyping services could substantially
method to accelerate the rate of genetic gain. reduce the genotyping costs, enabling the screening of a larger
number of accessions at the seedling stage. Development of
novel breeding strategies or models that not only capture
QTL UTILIZATION IN CASSAVA additive effects, but also account for dominance and epistatic
interactions, could contribute to the introgression of the QTLs
BREEDING into desired varieties. Likewise, the model should account for
Numerous QTL controlling a wide variety of traits have been QTL-environment interactions for an effective MAS scheme
identified to be utilized in marker-assisted selection (MAS) (Singh et al., 2019). Recent genomic innovations have prompt for
known as forward selection (Tables 2, 3). MAS is a technique of the search of new tools for population improvement.
indirect selection of traits. Its implementation is still challenging
in many breeding programs (Chukwu et al., 2019; Cobb et al.,
2019). Although panels of molecular markers closely linked GENOMIC SELECTION AND IT’s
to key cassava traits have been identified, successful applied POTENTIAL IN CASSAVA BREEDING
experiences of MAS in cassava breeding are limited (Table 4).
Markers tightly associated with CMD2 have been effectively used Genomic selection (GS), also referred to as genomic prediction,
to identify genotypes bearing the CMD2 gene in West African can complement MAS; with MAS being used for highly heritable
germplasm, introgression of CMD2 genes into various cassava traits controlled by one or a few markers, and GS being
germplasm, and selection of parental lines for planned crosses. used for more quantitative traits. GS relies on predicting the
This has enabled breeders to focus on fewer genotypes at an genomic estimated breeding values (GEBV) of an individual
early stage of the breeding program, saving the breeder time using a trait specific model built by simultaneously fitting
and labor cost (Blair et al., 2007; Okogbenin et al., 2012; do information provided by thousands of molecular markers spread
Carmo et al., 2015). MAS is appropriate for moderate to highly throughout the genome (Meuwissen et al., 2001). Fitting all
inherited traits that are controlled by a few major genes and not markers simultaneously allows a substantial fraction of trait
sufficiently predictive for quantitatively inherited traits controlled heritability missed by QTLs or association mapping to be
by many genes at different loci, each contribution to a small captured. These are likely to be small effect alleles (Resende
effect of the phenotypic expression (Singh et al., 2019). The et al., 2012). The approach was first championed in dairy cattle
time and investment required to develop reliable markers have (Jonas and de Koning, 2013). There are great expectations for
also been a limiting factor for MAS implementation in cassava the use of GS in cassava breeding. Since the first GS studies
breeding programs in SSA. Likewise, the use of convenient in cassava conducted by de Oliveira et al. (2012) and Ly et al.
phenotyping proxies do not always translate into meaningful (2013), highlighting the potential of GS, several studies have
selection targets for the breeding programs (Cobb et al., 2019). been published (Wolfe et al., 2017; Ozimati et al., 2018; de
Future endeavors should prioritize cassava traits for marker Andrade et al., 2019; Yonis et al., 2020). GS has been utilized
development using a stage-gate system (SGS) to manage trait in cassava breeding to increase resistance against CMD and
development. This will require the development of well-defined CBSD in cassava populations (Wolfe et al., 2016; Kayondo et al.,
cassava product profiles, a set of targeted attributes the new 2018; Ozimati et al., 2018). Ozimati et al. (2018) highlighted
variety should meet (Ragot et al., 2018). The establishment the potential of GS as a pre-emptive breeding strategy. Torres
of a SGS will ensure accountability, transparency, and data- et al. (2019) predicted good progress in selecting clones for traits
driven advancement decisions. This strategy is currently being such as fresh root yield (FRY), dry matter content (DMC), fresh
implemented with the support of Excellence in Breeding (EiB)1. shoot yield (FSY), harvest index (HI), dry yield (DY) using GS.
MAS should be considered only when proven more efficient than Yonis et al. (2020) assessed genomic prediction for root size
phenotypic selection to justify the expenses of development and and shape based on root traits extracted from digital images.
deployment of the locus in a breeding program. Likewise, marker Greater predictive ability has been reported for DMC, CMD,
development for the traits retained should be feasible in a short and, to a lesser extent, HI compared to other traits (Wolfe
to medium term to avoid waste of resources (Collard and Mackill, et al., 2016; de Andrade et al., 2019). This success has been
2008). The International Institute of Tropical Agriculture (IITA) attributed to their high to moderate heritability, large-effect
and partners have designed and converted molecular markers QTLs and low genotype × environment interaction (Torres
for several major gene traits, including provitamin A, CGM, et al., 2019; Yonis et al., 2020). GS is especially attractive for
complex traits controlled by many QTLs with low heritability
that are difficult or expensive to assess or are measured late in
1https://excellenceinbreeding.org/blog/applying-stage-gates-better-manage-
public-breeding-programs 2https://excellenceinbreeding.org/module3/kasp
Frontiers in Genetics | www.frontiersin.org 12 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 13
Mbanjo et al. Innovations for Improving Cassava Production
Frontiers in Genetics | www.frontiersin.org 13 January 2021 | Volume 11 | Article 623736
TABLE 4 | Key cassava traits, target regions, potential for GS and/or MAS and the current stage of trait development.
Traits Regiona Priorityb Method of evaluation Potential for MASc Stage 1d Stage 2e Stage 3f Stage 4g
Cassava mosaic disease All High Phenotyping/MAS/GS High X
Cassava brown streak East Africa High Phenotyping/GS Medium X
Cassava green mite All Medium Phenotyping/MAS/GS Medium X
White flies All High Phenotyping Undetermined X
Pro-vitamin A All High Phenotyping/MAS High
Low cyanogenic potential East Africa Medium Phenotyping High
Root mealiness East Africa and Central Africa Medium Phenotyping High
Dry matter content All High Phenotyping/GS Medium X
Starch content All High Phenotyping (low throughput) Medium X
Garri yield West Africa High Phenotyping (low throughput) Medium X
Fufu yield West Africa High Phenotyping (low throughput) Medium X
Plant type All High Phenotyping/GS Medium X
Harvest index All High Phenotyping/GS Low X
Root weight All High Phenotyping/GS Low X
Root number All High Phenotyping/GS Low X
Root size All High Phenotyping/GS Low X
aTarget regions; btrait priority; cpotential for marker-assisted selection; dProduct scoping; ePhenotyping and donor discovery; f locus identification; gLocus deployment; GS, Genomic selection; MAS, Marker-assisted
selection.
fgene-11-623736 January 18, 2021 Time: 16:34 # 14
Mbanjo et al. Innovations for Improving Cassava Production
the breeding cycle (de Oliveira et al., 2012). It could drastically compare different selection methods; (2) define optimal crossing
reduce the breeding cycle by choosing new parents based on strategies; and (3) determine appropriate recycling times. The
GEBV rather than actual phenotypes and limiting the size data intensive nature of GS and widespread use of this approach
and number of field experiments. However, refined strategies across more breeding programs requires development of shared
should be adopted for complex traits. Non-additive genetic computational infrastructures and analysis pipelines.
variation prevails for low heritability cassava traits and should
be accounted for. Therefore, models that capture non-additive
effects should be applied (Wolfe et al., 2016). GS, like any DEVELOPMENT OF COMPUTATIONAL
other approaches, is facing various challenges. The GS models TOOLS FOR A GREATER IMPACT
predict poorly across populations, and consequently, the strategy
requires continuous re-calibration with every breeding cycle Breeding modernization will necessitate the development of
(Wolfe et al., 2017). Allele frequency changes, introgression of robust statistical tools and innovative analytical pipelines and
new alleles, SNP-QTL linkage disequilibrium association, lack of platforms to accommodate the enormous quantity of data,
relatedness between germplasm and population structure, as well which is being generated. Due to its clonal propagation
as genotype-by-environment interaction effects, compromise method, there is a narrow timeline between harvesting and
the efficiency of GS. To overcome some of these limitations, the next cassava planting season. This necessitates quick
dual-purpose population development and variety development decision making and effective data management. Information
pipelines have been applied to ensure training data are closely systems are required to track samples, store genomic and
related to the new clones and for continuously updating the phenotypic information, merge data, and conduct analysis
training model (Santantonio et al., 2020). GS could be more to guide decision making (Santantonio et al., 2020). It is
robust by integrating biological knowledge. The inclusion of from this perspective that an open access repository, such
QTL markers associated with the trait of interest could increase as Cassavabase5, has been developed to centralize information
the robustness of genetic evaluation (Ozimati et al., 2018; Lan access to cassava research data and support cassava breeding
et al., 2020). Other variables affecting the precision of the programs (Fernandez-Pozo et al., 2015). Cassavabase contains
prediction model include the size of the training population, phenotypic, pedigree, and genomics data generated over
the number of markers used in the model, the trait genetic 30 years by cassava programs in Africa, Asia, and South
architecture, and heritability (de Oliveira et al., 2012; Ly et al., America. Cassavabase also encompasses computational tools to
2013; Wolfe et al., 2017; Somo et al., 2020). Poor predictions have facilitate analyses.
been reported across breeding programs limiting the prospect
of sharing data from different locations, breeding programs,
and countries (Wolfe et al., 2017; Somo et al., 2020). An GENOME EDITING ADVANCES AND
international project funded by UK Foreign, Commonwealth and PROSPECTS
Development Office (FCDO) and the Bill and Melinda Gates
Foundation (Gates Foundation) named the Next Generation Genome-editing is becoming a popular molecular tool for
Cassava Breeding Project3 is currently using this approach in functional genomics, as well as crop improvement. Clustered
four African research institutes, including the National Crop regularly interspaced palindromic repeats (CRISPR/CRISPR-
Resources Research Institute (NaCRRI) in Uganda, the National associated protein 9 (Cas9)-mediated genome-editing has rapidly
Root Crops Research Institute (NRCRI) in Nigeria, the Tanzania become the most popular genome engineering approach due
Agriculture Research Institute (TARI) in Tanzania, and IITA to its simplicity, efficiency, specificity, multiplexing and ease
in Nigeria. A multi-trait genomic selection strategy is being to adapt. Most of the CRISPR/Cas9-based genome-editing is
applied using program-based selection indices to efficiently reported for seed crops; however, recently, it is also established
improve quantitative traits simultaneously. The selection is for clonally propagated crops such as potato, banana, and
based on yield, DMC, virus resistance to CMD and CBSD, cassava (Butler et al., 2016; Odipio et al., 2017; Kaur et al.,
and good plant architecture. Outstanding clones are currently 2018; Naim et al., 2018; Tripathi et al., 2019; Ntui et al.,
advanced toward the variety release pipeline. The adoption of 2020). The accessibility of the genetic transformation system
GS has guided the mating design, enabling rapid pyramiding and reference genomes for cassava made it possible to realize
of favorable allele combinations and development of progeny the potential of CRISPR-based genome-editing for basic and
with the improved allelic combinations (de Oliveira et al., applied research to improve economically important cassava
2012). Moreover, GS is being used to increase genetic gain traits. The CRISPR/Cas9-based genome-editing system was
by decreasing the breeding cycle time, increasing selection demonstrated by knocking out the Phytoene desaturase (MePDS)
accuracy, and increasing selection intensity in early generations. gene in two cultivars of cassava, TMS60444 and TME204
In order to identify a more effective breeding strategy and (Odipio et al., 2017). Genome-editing was further applied
further improve breeding efficiency, computational simulations for developing a cassava variety with resistance against two
are currently being performed with the support of EIB4 to: (1) species of Ipomovirus, cassava brown streak virus (CBSV)
and Ugandan cassava brown streak virus (UCBSV), causing
3NextGen cassava, https://www.nextgencassava.org
4https://excellenceinbreeding.org/module2 5https://www.cassavabase.org/
Frontiers in Genetics | www.frontiersin.org 14 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 15
Mbanjo et al. Innovations for Improving Cassava Production
CBSD (Gomez et al., 2019). The targeted mutations in the collaborative projects involving international institutions,
translation initiation factor 4E (eIF4E) isoforms nCBP-1 and NARS, along with substantial funding from the donor
nCBP-2 in the edited cassava variety showed a reduction communities, have contributed to the increased number
in disease severity and virus accumulation in the storage of genome-wide association studies. Several gene(s)/QTLs
tuberous roots upon glasshouse challenge of edited cassava underlying key cassava traits have been identified and
lines with CBSV. Even though the mutations in the nCBP- trait-linked markers, a pre-requisite for MAS have been
1 and nCBP-2 genes conferred enhanced resistance to CBSD, developed (Ogbonna et al., 2020; Rabbi et al., 2020). The
complete resistance was not obtained. This suggests that total effective implementation of MAS hampered by economic
resistance to CBSD can be developed by stacking the genome- obstacles is currently being addressed with the support of
editing approach of disrupting eIF4E isoforms with other the EIB platform2, who seeks to mainstream the use of
resistance strategies such as RNAi (Gomez et al., 2019). Later, genotyping data. The platform through subsidies from the
researchers have attempted to apply this technology to develop Gates Foundation offers small breeding programs access
resistance against a geminivirus, African cassava mosaic virus to high quality genotyping services, including low-density
(ACMV) (Mehta et al., 2019). However, the edited cassava (Kompetitive allele specific PCR (KASP)) and mid-density
plants did not show significant resistance against ACMV (DArTAg) genotyping to foster the progressive integration
in greenhouse inoculation experiments. It might be due to of MAS into NARS crop breeding programs. The EIB low-
the evolution of editing-resistant geminiviruses in genome- density genotyping service is being used by IITA and NARS
edited cassava. CRISPR/Cas9 based genome-editing can be partners in Uganda, Tanzania, and Nigeria for quality control,
coupled to genetic improvements in cassava for traits such identification of cassava accessions with the desired alleles,
as starch improvement and early flowering. Cassava roots and validation of trait-markers. Genomic selection-based
normally produce large quantities of starch, having high amylose pilot projects are ongoing in Nigeria, Uganda, and Tanzania
levels, a crystallizable component that is more soluble in and the appealing perspectives offered by this approach,
water. However, the starch with low amylose levels, known including shortening of the breeding cycle and speed-up of
as “waxy starch,” is preferred for food processing and other variety development have been highlighted (Wolfe et al., 2017;
industrial uses. Bull et al. (2018) reported the application Kayondo et al., 2018; Somo et al., 2020). The mid-density
of CRISPR/Cas9 for manipulating starch biosynthesis and genotyping, DArTAg, a target genotyping provided by Diversity
improving the starch quality in the storage. They generated Arrays is being used as an alternative to GBS for genomic
edited cassava plants with mutations in two genes: protein selection applications. Although traditional methodologies
targeting to starch (PTST1) or granule bound starch synthase are still widely used for trait phenotyping, integration of high
(GBSS) involved in amylose biosynthesis, leading to reduction throughput phenotyping is on course. NIRS spectroscopy is
or elimination of amylose content. This, in turn, can improve being explored by few NARS programs to predict some key
the quality of starch cassava roots for commercial use. The cassava traits (i.e., carotenoids and DMC) and promising
authors also demonstrated accelerated breeding by transferring results have been reported (Ikeogu et al., 2017, 2019). NIRS
the Arabidopsis FLOWERING LOCUS T gene in the genome- evaluation and optimization for other traits, including gari,
edited events of cassava for early flowering. Genome-editing can fufu, starch, and root mealiness is ongoing. In collaboration
multiplex the traits, and researchers can develop cassava varieties with IDS GeoRadar6, IITA is testing a prototype commercial
with the waxy starch and early flowering. Despite still being GPR for routine cassava root phenotyping. The unmanned
in its infancy, genome-editing offers promising prospects for aerial vehicle (UAV) is currently used at IITA by the Cassava
cassava improvement and could shorten the breeding process. Source-Sink Project to quantify aboveground plant growth
Novel plant varieties could be directly used for crop production (Sonnewald et al., 2020). Image recognition has been evaluated
or as pre-breeding materials (Xu et al., 2019). Technological for high accuracy disease detection and mobile, as well as
developments like this should be followed by their adoption to web-based tools developed for cassava disease scoring and
increase productivity. monitoring7. An application programming interface has
been implemented within Cassavabase in order to process
cassava root images. Breeders can upload cassava root necrosis
CURRENT STATUS OF TECHNOLOGY sectional image captured during harvest and the result will
APPLICATIONS IN SUB-SAHARAN be returned back to Cassavabase. IITA and NARS partners
AFRICA have digitized all processes; data are uploaded, processed,
and accessed within Cassavabase to minimize human errors.
Effective implementation of technological development is CRISPR/Cas9-based genome-editing is mainly driven by
key to enhancing the productivity and profitability of cassava IITA, in partnership with a handful of national research
on the continent. Genetic studies conducted by national organizations. The difficulty in acquiring laboratory supplies,
institutes or academia in sub-Saharan Africa have been mainly as well as the need for constant and sufficient level of funding,
focused on germplasm characterization, genetic diversity constitute a bottleneck for the implementation of gene-editing
assessment, varietal identification, linkage mapping, and classical
QTL mapping (Fregene et al., 1997; Kawuki et al., 2009; 6https://idsgeoradar.com/
Rabbi et al., 2015; Adjebeng-Danquah et al., 2020). Recently, 7http://www.mcrops.org/
Frontiers in Genetics | www.frontiersin.org 15 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 16
Mbanjo et al. Innovations for Improving Cassava Production
technology by NARS. Likewise, the legal framework and CONCLUSION AND PERSPECTIVES
appropriate regulatory structures needed to guide the use of
this technology are still lacking in several African countries, It takes several years to develop improved cassava varieties.
hampering the movement of plants from laboratory to the field Genomic resources have facilitated the evaluation of local and
(Tripathi et al., 2020). Furthermore, biosafety considerations regional genetic diversity and profiling of breeding materials.
remain a public concern. Winning consumer’s acceptance and New phenotyping approaches are substituting traditional trait
trust is key to the effective implementation of this new evaluation and have been used for marker-trait associations,
breeding technique and to harness its full potential. Few cassava enabling identification of numerous QTLs for key traits.
breeding programs have benefited and/or explored these recent Although notable signs of progress have been achieved, especially
technological developments. For greater impact, the knowledge for less complex traits, genetic architecture is still not fully
and technologies outlined here should be disseminated and understood. For quantitatively inherited traits, phenotypic
transferred to other cassava breeding programs on the continent. plasticity and the difficult phenotypic evaluation complicates
matters further. There is a need to scale up multi-environment
evaluation. The development of advanced high throughput
TECHNOLOGY TRANSFER AND and accurate cassava phenotyping approaches is imperative.
CAPACITY DEVELOPMENT FOR MORE Translating those results obtained into practical breeding
SUSTAINABLE CASSAVA PRODUCTION methodologies and coherent biological knowledge is needed.
QTL results should be exploited, and validated trait-markers
Sustained research and innovation capacity is imperative for developed. Priorities will have to be set to ensure return on
agricultural transformation in Africa (Ojijo et al., 2016). Most investment. Therefore, establishment of a formal advancement
African national agricultural research systems (NARS) do not system with well-defined metrics is needed. Genetic gain should
have well-funded cassava breeding programs with sufficient be routinely monitored. Strategies need to be constantly revised
technical critical mass. Many programs routinely evaluate and as priorities evolve, and new challenges emerge. MAS and GS
multiply clones imported from larger breeding programs (i.e., are complementary breeding approaches that should be used in
from IITA or CIAT). Although most NARS would maintain tandem. More attention should be given to quality traits, which
certain capacity to develop and release varieties adapted to local have received less attention; whereas quality traits influence
agro-ecologies and local preferences, their breeding capacities varietal adoption and product utilization. Key quality traits
need to be strengthened and modernized to be effective. Many need to be defined and translated into biochemical parameters.
of the technological innovations are new to NARS and access to Advanced technologies, such as genome-editing, could play
equipment, reagents, and skilled personnel is challenging (Tester a prominent role in cassava improvement. However, further
and Langridge, 2010). It is with this in mind that an initiative such investigation will be required to ensure maximum benefits.
as the Cassava Community of Practice and Partnership (CoPP) It will be important that the new technologies and tools
has been established through the NextGen Cassava Project to developed are used by NARS. Therefore, regional networks,
serve as a platform to disseminate and facilitate the transfer of shared expertise, and service will be of prime importance. Last,
proven tools, methods, technologies, and products. The CoPP but not the least, the support and involvement of national
provides the initial technical backstopping, fosters collaboration, and regional governments is crucial for sustainable cassava
facilitates connectivity, and creates opportunities for peer-to-peer production on the continent.
learning between members. This will enable the establishment
of best practices and procedures and implementation of
recent technological innovations by NARS, increase breeding AUTHOR CONTRIBUTIONS
efficiencies, and create a broader and deeper impact. Successful
integration and application of innovative technologies necessitate EN conceived this review and drafted the manuscript. CE
technical expertise. Scaling up the research capabilities of NARS, conceived the idea and edited the manuscript. LT drafted a
through training, will hone their skills and knowledge (Bull et al., section of the manuscript. MF, IR, EH, KSI, PK, and CE provided
2011). The CoPP concept is not new. A successful example is critical edits. All authors contributed to the article and approved
the sweet potato breeding Community of Practice, which has the submitted version.
strengthened national capacities leading to the development and
released of several user-preferred varieties with impact on the
quality of life of small farmers (SASHA, 2019). The cassava FUNDING
CoPP will need to evolve into a cohesive network with clear
accountabilities and expectations. NARS and CGIAR will need to The authors thank the UK’s Foreign, Commonwealth and
work together in a coordinated breeding network for economies Development Office (FCDO) and the Bill and Melinda Gates
of scale. Government involvement will be needed to sustain the Foundation (Grant INV-007637, http://www.gatesfoundation.
gain achieved by the breeders. org) for their financial support.
Frontiers in Genetics | www.frontiersin.org 16 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 17
Mbanjo et al. Innovations for Improving Cassava Production
REFERENCES to African laboratories. Plant Cell Rep. 30, 779–787. doi: 10.1007/s00299-010-
0986-6
Abdullakasim, W., Powbunthorn, K., Unartngam, J., and Takigawa, T. (2011). An Bull, S. E., Seung, D., Chanez, C., Mehta, D., Kuon, J. E., Truernit, E., et al. (2018).
images analysis technique for recognition of brown leaf spot disease in cassava. Accelerated ex situ breeding of GBSS- and PTST1-edited cassava for modified
Tarım Makinaları Bilim. Derg. 7, 165–169. starch. Sci. Adv. 4:eaat6086. doi: 10.1126/sciadv.aat6086
Adinsi, L., Akissoé, N., Escobar, A., Prin, L., Kougblenou, N., Dufour, D., et al. Butler, N. M., Baltes, N. J., Voytas, D. F., and Douches, D. S. (2016).
(2019). Sensory and physicochemical profiling of traditional and enriched gari Geminivirus-mediated genome editing in potato (Solanum tuberosum l.) using
in Benin. Food Sci. Nutr. 00, 1–11. doi: 10.1002/fsn3.1201 sequence-specific nucleases. Front. Plant Sci. 7:1045. doi: 10.3389/fpls.2016.
Adjebeng-Danquah, J., Manu-Aduening, J., Asante, I. K., Agyare, R. Y., Gracen, 01045
V., and Offei, S. K. (2020). Genetic diversity and population structure analysis Carvalho, L. M. J., Oliveira, A. R. G., Godoy, R. L. O., Pacheco, S., Nutti, M. R., De
of Ghanaian and exotic cassava accessions using simple sequence repeat (SSR) Carvalho, J. L. V., et al. (2012). Retention of total carotenoid and β-carotene in
markers. Heliyon 6:e03154. doi: 10.1016/j.heliyon.2019.e03154 yellow sweet cassava (Manihot esculenta Crantz) after domestic cooking. Food
Afonso, T., Moresco, R., Uarrota, V. G., Navarro, B. B., da Nunes, E. C. N., Nutr. Res. 56:15788. doi: 10.3402/fnr.v56i0.15788
Maraschin, M., et al. (2017). UV-Vis and CIELAB based chemometric Ceballos, H., Pérez, J. C., Barandica, O. J., Lenis, J. I., Morante, N., Calle, F., et al.
characterization of Manihot esculenta carotenoid contents∗. J. Integr. Bioinform. (2016). Cassava breeding I: The value of breeding value. Front. Plant Sci. 7:1227.
14:20170056. doi: 10.1515/jib-2017-0056 doi: 10.3389/fpls.2016.01227
Akano, A. O., Dixon, A. G. O., Mba, C., Barrera, E., and Fregene, M. (2002). Genetic Chatpapamon, C., Wandee, Y., Uttapap, D., Puttanlek, C., and Rungsardthong,
mapping of a dominant gene conferring resistance to cassava mosaic disease. V. (2019). Pasting properties of cassava starch modified by heat-moisture
Theor. Appl. Genet. 105, 521–525. doi: 10.1007/s00122-002-0891-7 treatment under acidic and alkaline pH environments. Carbohydr. Polym. 215,
Akely, P. M. T., Djina, Y., Konan, B. R., Irie, K., Kouame, L. P., and Amani, 338–347. doi: 10.1016/j.carbpol.2019.03.089
N. G. (2016). Study of Varietal Influence Post Conservation on Biochemical and Chiang, K. S., Bock, C. H., El Jarroudi, M., Delfosse, P., Lee, I. H., and Liu,
Sensory Qualities of Attiéké and Boiled Cassava (Manihot esculenta Crantz). H. I. (2016). Effects of rater bias and assessment method on disease severity
Agric. Sci. 7, 127–136. doi: 10.4236/as.2016.73012 estimation with regard to hypothesis testing. Plant Pathol. 65, 523–535. doi:
Akinbo, O., Gedil, M., Ekpo, E. J. A., Oladele, J., and Dixon, A. G. O. (2007). 10.1111/ppa.12435
Detection of RAPD markers-linked to resistance to cassava anthracnose disease. Chisenga, S. M., Workneh, T. S., Bultosa, G., and Laing, M. (2019).
African J. Biotechnol. 6, 677–682. doi: 10.5897/AJB2007.000-2068 Characterization of physicochemical properties of starches from improved
Akinpelu, A. O., Amamgbo, L. E. F., Olojede, A. O., and Oyekale, A. S. (2011). cassava varieties grown in Zambia. AIMS Agric. Food 4, 939–966. doi: 10.3934/
Health implications of cassava production and consumption. J. Agric. Soc. Res. agrfood.2019.4.939
11, 118–125. Choperena, E. D. P. M., Ospina, C., Fregene, M., Montoya-Lerma, J., and
Alicai, T., Szyniszewska, A. M., Omongo, C. A., Abidrabo, P., Okao-Okuja, G., Bellotti, A. C. (2012). Identificación de microsatélites en yuca asociados con la
Baguma, Y., et al. (2019). Expansion of the cassava brown streak pandemic in resistencia al ácaro Mononychellus tanajoa (acari: Tetranychidae). Rev. Colomb.
Uganda revealed by annual field survey data for 2004 to 2017. Sci. Data 6:327. Entomol. 38, 70–75.
doi: 10.1038/s41597-019-0334-9 Chueamchaitrakun, P., Chompreeda, P., Haruthaithanasan, V., Suwonsichon, T.,
An, O., Uj, U., and Re, K. (2019). Physicochemical properties of cassava processing and Kasemsamran, S. (2011). Prediction of pasting and thermal properties of
residue flour and sensory evaluation of fufu prepared from it. J. Nutraceuticals mixed Hom-Mali and glutinous rice flours using near infrared spectroscopy.
Food Sci. 4, 1–6. Kasetsart J. Nat. Sci. 45, 481–489.
Anderson, B., Eghan, M. J., Asare-Bediako, E., and Buah-Bassuah, P. K. (2015). Chukwu, S. C., Rafii, M. Y., Ramlee, S. I., Ismail, S. I., Oladosu, Y., Okporie, E.,
Optical imaging method for determining symtoms severity of cassava mosaic et al. (2019). Marker-assisted selection and gene pyramiding for resistance to
disease. Appl. Phys. Res. 7:34. bacterial leaf blight disease of rice (Oryza sativa L.). Biotechnol. Biotechnol.
Ayetigbo, O., Latif, S., Abass, A., and Müller, J. (2018). Comparing characteristics Equip. 33, 440–455. doi: 10.1080/13102818.2019.1584054
of root, flour and starch of biofortified yellow-flesh and white-flesh cassava Cobb, J. N., Biswas, P. S., and Platten, J. D. (2019). Back to the future: revisiting
variants, and sustainability considerations: A review. Sustain 10:3089. doi: 10. MAS as a tool for modern plant breeding. Theor. Appl. Genet. 132, 647–667.
3390/su10093089 doi: 10.1007/s00122-018-32663264
Bechoff, A., Tomlins, K., Fliedel, G., Becerra Lopez-lavalle, L. A., Westby, A., Collard, B. C. Y., Jahufer, M. Z. Z., Brouwer, J. B., and Pang, E. C. K. (2005). An
Hershey, C., et al. (2018). Cassava traits and end-user preference: Relating traits introduction to markers, quantitative trait loci (QTL) mapping and marker-
to consumer liking, sensory perception, and genetics. Crit. Rev. Food Sci. Nutr. assisted selection for crop improvement: The basic concepts. Euphytica 142,
58, 547–567. doi: 10.1080/10408398.2016.1202888 169–196. doi: 10.1007/s10681-005-1681-5
Belalcazar, J., Dufour, D., Andersson, M. S., Pizarro, M., Luna, J., Londoño, L., Collard, B. C., and Mackill, D. J. (2008). Marker-assisted selection: an approach for
et al. (2016). High-throughput phenotyping and improvements in breeding precision plant breeding in the twenty-first century. Philos. Trans. R. Soc. B Biol.
cassava for increased carotenoids in the roots. Crop Sci. 56, 2916–2925. doi: Sci. 363, 557–572. doi: 10.1098/rstb.2007.2170
10.2135/cropsci2015.11.0701 Cozzolino, D., Roumeliotis, S., and Eglinton, J. (2013). Relationships between
Blair, M. W., Fregene, M. A., Beebe, S. E., and Ceballos, H. (2007). “Marker- swelling power, water solubility and near-infrared spectra in whole grain barley:
assisted selection in common beans and cassava,” in Marker-assisted selection: A feasibility study. Food Bioprocess Technol. 6, 2732–2738. doi: 10.1007/s11947-
current status and future perspectives in crops, livestock, forestry and fish, eds E. 012-0948-9
Guimarães, J. Ruane, B. Scherf, A. Sonnino, and J. Dargie (Rome: FAO), 81–115. de Andrade, L. R. B., Bandeirae Sousa, M., Oliveira, E. J., de Resende, M. D. V.,
Boonpo, S., and Kungwankunakorn, S. (2017). Study on amylose iodine complex and Azevedo, C. F. (2019). Cassava yield traits predicted by genomic selection
from cassava starch by colorimetric method. J. Adv. Agric. Technol. 4, 345–349. methods. PLoS One 14:e0224920. doi: 10.1371/journal.pone.0224920
doi: 10.18178/joaat.4.4.345-349 de Oliveira, E. J., de Resende, M. D. V., da Silva, Santos, V., Ferreira, C. F.,
Bredeson, J. V., Lyons, J. B., Prochnik, S. E., Wu, G. A., Ha, C. M., Edsinger- Oliveira, G. A. F., et al. (2012). Genome-wide selection in cassava. Euphytica
Gonzales, E., et al. (2016). Sequencing wild and cultivated cassava and related 187, 263–276. doi: 10.1007/s10681-012-0722-0
species reveals extensive interspecific hybridization and genetic diversity. Nat. Delgado, A., Hays, D. B., Bruton, R. K., Ceballos, H., Novo, A., Boi, E., et al. (2017).
Biotechnol. 34, 562–571. doi: 10.1038/nbt.3535 Ground penetrating radar: A case study for estimating root bulking rate in
Brito, A. C., Oliveira, S. A. S., and Oliveira, E. J. (2017). Genome-wide association cassava (Manihot esculenta Crantz). Plant Methods 13:65. doi: 10.1186/s13007-
study for resistance to cassava root rot. J. Agric. Sci. 155, 1424–1441. doi: 017-0216-0
10.1017/S0021859617000612 Delgado, A., Novo, A., and Hays, D. B. (2019). Data acquisition methodologies
Bull, S. E., Ndunguru, J., Gruissem, W., Beeching, J. R., and Vanderschuren, H. utilizing ground penetrating radar for cassava (Manihot esculenta Crantz) root
(2011). Cassava: Constraints to production and the transfer of biotechnology architecture. Geosci 9:171. doi: 10.3390/geosciences9040171
Frontiers in Genetics | www.frontiersin.org 17 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 18
Mbanjo et al. Innovations for Improving Cassava Production
do Carmo, C. D., da Silva, M. S., Oliveira, G. A. F., and de Oliveira, E. J. (2015). nCBP-1 and nCBP-2 reduces cassava brown streak disease symptom severity
Molecular-assisted selection for resistance to cassava mosaic disease in Manihot and incidence. Plant Biotechnol. J. 17, 421–434. doi: 10.1111/pbi.12987
esculenta Crantz. Sci. Agric. 72, 520–527. doi: 10.1590/0103-9016-2014-0348 ICGMC. (2015). High-resolution linkage map and chromosome-scale genome
do Carmo, C. D., e Sousa, M. B., Brito, A. C., and de Oliveira, E. J. (2020). assembly for cassava (Manihot esculenta crantz) from 10 populations. G3 Genes
Genome-wide association studies for waxy starch in cassava. Euphytica 216:82. Genomes Genet 5, 133–144. doi: 10.1534/g3.114.015008
doi: 10.1007/s10681-020-02615-9 Iglesias, C., Mayer, J., Chavez, L., and Calle, F. (1997). Genetic potential and
Donkor, E., Onakuse, S., Bogue, J., de Los Rios, and Carmenado, I. (2017). The stability of carotene content in cassava roots. Euphytica 94, 367–373. doi: 10.
impact of the presidential cassava initiative on cassava productivity in Nigeria: 1023/A:1002962108315
Implication for sustainable food supply and food security. Cogent Food Agric. Ikeogu, U. N., Akdemir, D., Wolfe, M. D., Okeke, U. G., Chinedozi, A., Jannink,
3:1368857. doi: 10.1080/23311932.2017.1368857 J. L., et al. (2019). Genetic correlation, genome-wide association and genomic
Echefu, S. U., Rabbi, I. Y., Nwachukwu, E. C., and Adamu, C. L. Y. (2016). Mapping prediction of portable nirs predicted carotenoids in cassava roots. Front. Plant
of Quantitative Trait Loci Associated With Cassava Mosaic Disease (Cmd) Sci. 10:1570. doi: 10.3389/fpls.2019.01570
Resistance. Int. J. Sci. Environ. 5, 2637–2645 Ikeogu, U. N., Davrieux, F., Dufour, D., Ceballos, H., Egesi, C. N., and Jannink, J.
Elshire, R. J., Glaubitz, J. C., Sun, Q., Poland, J. A., Kawamoto, K., Buckler, E. S., L. (2017). Rapid analyses of dry matter content and carotenoids in fresh cassava
et al. (2011). A robust, simple genotyping-by-sequencing (GBS) approach for roots using a portable visible and near infrared spectrometer (Vis/NIRS). PLoS
high diversity species. PLoS One 6:e19379. doi: 10.1371/journal.pone.0019379 One 12:e0188918.
Esuma, W., Kawuki, R. S., Herselman, L., and Labuschagne, M. T. (2016). Diallel Iliya, J. M., and Madumelu, M. (2019). Determination of cyanide conetent in
analysis of provitamin A carotenoid and dry matter content in cassava (Manihot cassava tubers (Manihot esculenta) and apple seed (Pyrus malus). Int. J. Curr.
esculenta Crantz). Breed. Sci. 66, 627–635. doi: 10.1270/jsbbs.15159 Res. Chem. Pharm. Sci. 6, 14–20. doi: 10.22192/ijcr
Ezenwaka, L., Carpio, D. P., Del, Jannink, J. L., Rabbi, I., Danquah, E., et al. (2018). Jaramillo, A. M., Londoño, L. F., Orozco, J. C., Patiño, G., Belalcazar, J., Davrieux,
Genome-wide association study of resistance to cassava green mite pest and F., et al. (2018). A comparison study of five different methods to measure
related traits in cassava. Crop Sci. 58, 1–12. doi: 10.2135/cropsci2018.01.0024 carotenoids in biofortified yellow cassava (Manihot esculenta). PLoS One.
Ezenwaka, L., Rabbi, I., Onyeka, J., Kulakow, P., and Egesi, C. (2020). Identification 13:e0209702. doi: 10.1371/journal.pone.\underbar{0209702}A ML. FJ. CGJF
of additional /novel QTL associated with resistance to cassava green mite in a Jonas, E., and de Koning, D. J. (2013). Does genomic selection have a future in plant
biparental mapping population. PLoS One 15:e0231008. doi: 10.1371/journal. breeding? Trends Biotechnol. 31, 497–504. doi: 10.1016/j.tibtech.2013.06.003
pone.0231008 Jorge, V., Fregene, M. A., Duque, M. C., Bonierbale, M. W., Tohme, J., and Verdier,
Fanou, A. A., Zinsou, V. A., and Wydra, K. (2018). Cassava bacterial blight: a V. (2000). Genetic mapping of resistance to bacterial blight disease in cassava
devastating disease of cassava. London: Intechopen, 14–36. (Manihot esculenta Crantz). Theor. Appl. Genet. 101, 865–872. doi: 10.1007/
FAOSTAT. (2020). http://www.fao.org/faostat/en/#data/QC (Accessed January 24, s001220051554
2020). Jorge, V., Fregene, M., Vélez, C. M., Duque, M. C., Tohme, J., and Verdier,
Feleke, S., Manyong, V., Abdoulaye, T., and Alene, A. D. (2016). Assessing the V. (2001). QTL analysis of field resistance to Xanthomonas axonopodis
impacts of cassava technology on poverty reduction in Africa. Stud. Agric. Econ. pv. manihotis in cassava. Theor. Appl. Genet. 102, 564–571. doi: 10.1007/
118, 101–111. doi: 10.7896/j.1612 s001220051683
Ferguson, M. E., Hearne, S. J., Close, T. J., Wanamaker, S., Moskal, W. A., Town, Kainuma, K., Odat, T., and Cuzuki, S. (1967). Study of starch phosphates
C. D., et al. (2012). Identification, validation and high-throughput genotyping monoesters. J. Tech. Soc. Starch 14, 24–28
of transcribed gene SNPs in cassava. Theor. Appl. Genet. 124, 685–695. doi: Kalyebi, A., MacFadyen, S., Parry, H., Tay, W. T., De Barro, P., and Colvin, J. (2018).
10.1007/s00122-011-1739-9 African cassava whitefly, Bemisia tabaci, cassava colonization preferences
Ferguson, M. E., Shah, T., Kulakow, P., and Ceballos, H. (2019). A global overview and control implications. PLoS One 13:e0204862. doi: 10.1371/journal.pone.
of cassava genetic diversity. PLoS One 14:e0224763. doi: 10.1371/journal.pone. 0204862
0224763 Kamanda, I., Blay, E. T., Asante, I. K., Danquah, A., Ifie, B. E., Parkes, E., et al.
Ferguson, M., Rabbi, I., Kim, D., Gedil, M., Lopez-Lavalle, L. A. B., (2020). Genetic diversity of provitamin-A cassava (Manihot esculenta Crantz)
and Okogbenin, E. (2011). Molecular Markers and their Application in Sierra Leone. Genet. Resour. Crop Evol. 67, 1193–1208. doi: 10.1007/s10722-
to Cassava Breeding: Past, Present and Future. Trop. Plant Biol. 5, 020-00905-8
95–109. Kaur, N., Alok, A. S., Kaur, N., Pandey, P., Awasthi, P., et al. (2018). CRISPR/Cas9-
Fernandez-Pozo, N., Menda, N., Edwards, J. D., Saha, S., Tecle, I. Y., Strickler, S. R., mediated efficient editing in phytoene desaturase (PDS) demonstrates precise
et al. (2015). The Sol Genomics Network (SGN)-from genotype to phenotype to manipulation in banana cv. Rasthali genome. Funct. Integr. Genom. 18, 89–99.
breeding. Nucleic Acids Res. 43, D1036–D1041. doi: 10.1093/nar/gku1195 doi: 10.1007/s10142-017-0577-5
Fregene, M., Angel, F., Gomez, R., Rodriguez, F., Chavarriaga, P., Roca, W., et al. Kawuki, R. S., Esuma, W., Ozimati, A., Kayondo, I. S., Nandudu, L., and Wolfe,
(1997). A molecular genetic map of cassava (Manihot esculenta crantz). Theor. M. (2019). Alternative approaches for assessing cassava brown streak root
Appl. Genet. 95, 431–441. necrosis to guide resistance breeding and selection. Front. Plant Sci. 10:1461.
FSIN (2019). Global report on food crises. Available online at: https://www.wfp.org/ doi: 10.3389/fpls.2019.01461
publications/2020-global-report-food-crises (Accessed on October 23, 2020) Kawuki, R. S., Ferguson, M., Labuschagne, M., Herselman, L., and Kim, D. J.
Fukuda, W. M. G., Guevara, C. L., Kawuki, R., and Ferguson, M. E. (2010). Selected (2009). Identification, characterisation and application of single nucleotide
morphological and agronomic descriptors for the characterization of cassava. polymorphisms for diversity assessment in cassava (Manihot esculenta Crantz).
Ibadan:, NigeriaIITA, 19. Mol. Breed. 23, 669–684. doi: 10.1007/s11032-009-9264-0
Fukushima, A. R., Nicoletti, M. A., Rodrigues, A. J., Pressutti, C., Almeida, J., Kayondo, S. I., Pino Del, Carpio, D., Lozano, R., Ozimati, A., Wolfe, M., et al.
Brandão, T., et al. (2016). Cassava flour: Quantification of cyanide content. Food (2018). Genome-wide association mapping and genomic prediction for CBSD
Nutr. Sci. 7, 592–599. doi: 10.4236/fns.2016.77060 resistance in Manihot esculenta. Sci. Rep. 8:1549. doi: 10.1038/s41598-018-
Garcia-Oliveira, A. L., Kimata, B., Kasele, S., Kapinga, F., Masumba, E., Mkamilo, 19696-1
G., et al. (2020). Genetic analysis and QTL mapping for multiple biotic Kengkanna, J., Jakaew, P., Amawan, S., Busener, N., Bucksch, A., and Saengwilai,
stress resistance in cassava. PLoS One 15:e0236674. doi: 10.1371/journal.pone. P. (2019). Phenotypic variation of cassava root traits and their responses to
0236674 drought. Appl. Plant Sci. 7:e1238. doi: 10.1002/aps3.1238
Gegios, A., Amthor, R., Maziya-Dixon, B., Egesi, C., Mallowa, S., Nungo, R., et al. Khandare, V., and Choomsook, P. (2019). Cassava export of Thailand: Growth
(2010). Children consuming cassava as a staple food are at risk for inadequate performance and composition. Int. J. Res. Anal. Rev. 6, 847–857.
zinc, iron, and vitamin A intake. Plant Foods Hum. Nutr. 65, 64–70. doi: 10. Kizito, E. B., Rönnberg-Wästljung, A.-C., Egwang, T., Gullberg, U., Fregene, M.,
1007/s11130-010-0157155 and Westerbergh, A. (2007). Quantitative trait loci controlling cyanogenic
Gomez, M. A., Lin, Z. D., Moll, T., Chauhan, R. D., Hayden, L., Renninger, K., et al. glucoside and dry matter content in cassava (Manihot esculenta Crantz) roots.
(2019). Simultaneous CRISPR/Cas9-mediated editing of cassava eIF4E isoforms Hereditas 144, 129–136. doi: 10.1111/j.2007.0018-0661.01975.x
Frontiers in Genetics | www.frontiersin.org 18 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 19
Mbanjo et al. Innovations for Improving Cassava Production
Koros, J. C., Runo, S. M., Yusuf, M., and Orek, C. O. (2018). Screening selected Naim, F., Dugdale, B., Kleidon, J., Brinin, A., Shand, K., Waterhouse, P., et al.
cassava cultivars for resistance against cassava viruses and cassava green mites (2018). Gene editing the phytoene desaturase alleles of Cavendish banana using
under advanced yield trials in Kenya. IOSR J. Biotechnol. Biochem. 4, 37–52. CRISPR/Cas9. Transgenic Res. 27, 451–460. doi: 10.1007/s11248-018-0083-0
doi: 10.9790/264X-0405023752 Nakatumba-Nabende, J., Akera, B., Tusubira, J. F., Nsumba, S., and Mwebaze,
Kretzschmar, T., Mbanjo, E. G. N., Magalit, G. A., Dwiyanti, M. S., Habib, E. (2020). A dataset of necrotized cassava root cross-section images. Data Br.
M. A., Diaz, M. G., et al. (2018). DNA fingerprinting at farm level maps rice 32:106170. doi: 10.1016/j.dib.2020.106170
biodiversity across Bangladesh and reveals regional varietal preferences. Sci. Nassar, N., and Ortiz, R. (2010). Breeding Cassava to Feed the Poor. Sci. Am. 302,
Rep. 8:14920. doi: 10.1038/s41598-018-33080-z 78–85. doi: 10.1002/9781118060995.ch3
Kuon, J. E., Qi, W., Schläpfer, P., Hirsch-Hoffmann, M., Von Bieberstein, P. R., Ni, J., Pujar, A., Youens-Clark, K., Yap, I., Jaiswal, P., Tecle, I., et al. (2009).
Patrignani, A., et al. (2019). Haplotype-resolved genomes of geminivirus- Gramene QTL database: Development, content and applications. Database
resistant and geminivirus-susceptible African cassava cultivars. BMC Biol. 2009:ba005. doi: 10.1093/database/bap005
17:75. doi: 10.1186/s12915-019-0697-6 Noranizan, M. A., Dzulkifly, M. H., and Russly, A. R. (2010). Effect of heat
Lan, S., Zheng, C., Hauck, K., McCausland, M., Duguid, S. D., Booker, H. M., treatment on the physico-chemical properties of starch from different botanical
et al. (2020). Genomic prediction accuracy of seven breeding selection traits sources. Int. Food Res. J. 17, 127–135.
improved by QTL identification in flax. Int. J. Mol. Sci. 21:1577. doi: 10.3390/ Ntui, V. O., Tripathi, J. N., and Tripathi, L. (2020). Robust CRISPR/Cas9 mediated
ijms21051577 genome editing tool for banana and plantain (Musa spp.). Curr. Plant Biol.
Leach, H. W. (1959). Structure of the starch granule I swelling and solubility 2020:100128. doi: 10.1016/j.cpb.2019.100128
patterns of various starches. Cereal Chem. 36, 534–544. Nzuki, I., Katari, M. S., Bredeson, J. V., Masumba, E., Kapinga, F., Salum, K., et al.
Lu, G., Huang, H., and Zhang, D. (2006a). Application of near-infrared (2017). QTL mapping for pest and disease resistance in cassava and coincidence
spectroscopy to predict sweetpotato starch thermal properties and noodle of some QTL with introgression regions derived from Manihot Glaziovii. Front.
quality. J. Zhejiang Univ. Sci. B. 7, 475–481. doi: 10.1631/jzus.2006.B0475 Plant Sci. 8:1168. doi: 10.3389/fpls.2017.01168
Lu, G., Huang, H., and Zhang, D. (2006b). Prediction of sweetpotato starch Odipio, J., Alicai, T., Ingelbrecht, I., Nusinow, D. A., Bart, R., and Taylor,
physiochemical quality and pasting properties using near-infrared reflectance N. J. (2017). Efficient CRISPR/cas9 genome editing of phytoene desaturase in
spectroscopy. Food Chem. 94, 632–639. doi: 10.1016/j.foodchem.2005.02.006 cassava. Front. Plant Sci. 8:1780. doi: 10.3389/fpls.2017.01780
Luo, X., Tomlins, K. I., Carvalho, L. J. C. B., Li, K., and Chen, S. (2018). The Ogbonna, A. C., de Andrade, L. R., Rabbi, I. Y., Mueller, L. A., de Oliveira, E. J.,
analysis of candidate genes and loci involved with carotenoid metabolism in and Bauchet, G. J. (2020). ). Genetic architecture and gene mapping of cyanide
cassava (Manihot esculenta Crantz) using SLAF-seq. Acta Physiol. Plant. 40:66. in cassava (Manihot esculenta Crantz). bioRxiv preprint doi: \url{https://org/10.
doi: 10.1007/s11738-018-2634-7 1101/2020.06.19.159160}
Ly, D., Hamblin, M., Rabbi, I., Melaku, G., Bakare, M., Gauch, H. G., et al. Ojijo, N., Franzel, S., Simtowe, F., Madakadze, R., Nkwake, A., and Moleko, L.
(2013). Relatedness and genotype × environment interaction affect prediction (2016). “The role for agricultural Research Systems, Advisory services and
accuracies in genomic selection: A study in cassava. Crop Sci. 53, 1312–1325. capacity developement and knowledge transfer,” in Africa agriculture status
doi: 10.2135/cropsci2012.11.0653 report 2016: progress towards agricultural transformation (AGRA). (Uttar
Ma’Aruf, A. G., and Abdul, H. R. (2020). Efficient processing of cassava starch: Pradesh: AGRA)200–230.
Physicochemical characterization at different processing parameters. Food Res. Okogbenin, E., Egesi, C. N., Olasanmi, B., Ogundapo, O., Kahya, S., Hurtado, P.,
4, 143–151. doi: 10.26656/fr.2017.4(1).235 et al. (2012). Molecular marker analysis and validation of resistance to cassava
Maieves, H. A., Oliveira, D. C., De, Bernardo, C., Müller, C. M. D. O., and Amante, mosaic disease in elite cassava genotypes in Nigeria. Crop Sci. 52, 2576–2586.
E. R. (2012). Microscopy and texture of raw and cooked cassava (Manihot doi: 10.2135/cropsci2011.11.0586
esculenta Crantz) roots. J. Texture Stud. 43, 164–173. doi: 10.1111/j.1745-4603. Okogbenin, E., Marin, J., and Fregene, M. (2006). An SSR-based molecular genetic
2011.00327.x map of cassava. Euphytica 147, 433–440
Malik, A. I, Kongsil, P., Nguyễn, V. A., Ou, W., and Srean, P. (2020). Cassava Okogbenin, E., Marin, J., and Fregene, M. (2008). QTL analysis for early yield in
breeding and agronomy in Asia: 50 years of history and future directions. Breed. a pseudo F2 population of cassava. Afr. J. Biotechnol. 7, 131–138. doi: 10.5897/
Sci. 70, 145–166. doi: 10.1270/jsbbs.18180 AJB2008.000-5017
Masumba, E. A., Kapinga, F., Mkamilo, G., Salum, K., Kulembeka, H., Rounsley, Otekunrin, O. A., and Sawicka, B. (2019). Cassava, a 21st century staple crop: How
S., et al. (2017). QTL associated with resistance to cassava brown streak and can Nigeria harness its enormous trade potentials? Acta Sci. Agric. 3, 194–202.
cassava mosaic diseases in a bi-parental cross of two Tanzanian farmer varieties, doi: 10.31080/asag.2019.03.0586
Namikonga and Albert. Theor. Appl. Genet. 130, 2069–2090. doi: 10.1007/ Owomugisha, G., and Mwebaze, E. (2016). Machine learning for plant disease
s00122-017-2943-z incidence and severity measurements from leaf images. in Proceedings - 2016
Mehta, D., Stürchler, A., Anjanappa, R. B., Zaidi, S. S. E. A., Hirsch-Hoffmann, 15th IEEE International Conference on Machine Learning and Applications.
M., Gruissem, W., et al. (2019). Linking CRISPR-Cas9 interference in cassava ICMLA 2016, 158–163. doi: 10.1109/ICMLA.2016.126
to the evolution of editing-resistant geminiviruses. Genome Biol. 20:80. doi: Ozimati, A., Kawuki, R., Esuma, W., Kayondo, I. S., Wolfe, M., Lozano, R.,
10.1186/s13059-019-1678-3 et al. (2018). Training population optimization for prediction of cassava brown
Meullenet, J. F., Mauromoustakos, A., Horner, T. B., and Marks, B. P. (2002). streak disease resistance in West African Clones. G3 Genes Genomes Genet. 8,
Prediction of texture of cooked white rice by near-infrared reflectance analysis 3903–3913. doi: 10.1534/g3.118.200710
of whole-grain milled samples. Cereal Chem. 79, 52–57. doi: 10.1094/CCHEM. Patil, B. L., Legg, J. P., Kanju, E., and Fauquet, C. M. (2015). Cassava brown
2002.79.1.52 streak disease: a threat to food security in Africa. J. Gen. Virol. 96, 956–968.
Meuwissen, T. H. E., Hayes, B. J., and Goddard, M. E. (2001). Prediction of total doi: 10.1099/jgv.0.000014
genetic value using genome-wide dense marker maps. Genetics 157, 1819–1829. Ping, H., Wang, J., and Ren, G. (2013). Prediction of the total starch and amylose
Moriasi, G. A., Olela, B. O., Waiganjo, B. W., Wakori, E. W. T., and Onyancha, content in barley using near-infrared reflectance spectroscopy. Intell. Autom.
J. M. (2017). Evaluation of cyanide levels in two cassava varieties (Mariwa and Soft. Comput. 19, 231–237. doi: 10.1080/10798587.2013.823719
Nyakatanegi) grown in Bar-agulu, Siaya County, Kenya. J. Food Nutr. Res. 5, Pootakham, W., Shearman, J. R., Ruang-Areerate, P., Sonthirod, C., Sangsrakru, D.,
817–823. doi: 10.12691/jfnr-5-1114 Jomchai, N., et al. (2014). Large-scale SNP discovery through RNA sequencing
Morillo, A. C. C., Morillo, Yacenia, C., and Ceballos, H. L. (2013). Identification of and SNP genotyping by targeted enrichment sequencing in cassava (Manihot
QTLs for carotene content in the genome of cassava (Manihot esculenta Crantz) esculenta Crantz). PLoS One 9:e116028. doi: 10.1371/journal.pone.011
and S1 population validation. Acta Agron. 62, 196–206. 6028
Mtunguja, M. K., Beckles, D. M., Laswai, H. S., Ndunguru, J. C., and Sinha, Prohens, J. (2011). Plant breeding: a success story to be continued thanks to the
N. J. (2019). Opportunities to commercialize cassava production for poverty advances in genomics. Front. Plant Sci. 2:51. doi: 10.3389/fpls.2011.00051
alleviation and improved food security in Tanzania. African J. Food, Agric. Nutr. Rabbi, I. Y., Hamblin, M. T., Kumar, P. L., Gedil, M. A., Ikpan, A. S., Jannink,
Dev. 19, 13928–13946. doi: 10.18697/AJFAND.84.BLFB1037 J. L., et al. (2014). High-resolution mapping of resistance to cassava mosaic
Frontiers in Genetics | www.frontiersin.org 19 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 20
Mbanjo et al. Innovations for Improving Cassava Production
geminiviruses in cassava using genotyping-by-sequencing and its implications Breeding-Community-Practice-Breeding-Africa-2009-2019-4p_web.pdf
for breeding. Virus Res. 186, 87–96. doi: 10.1016/j.virusres.2013.12.028 (accessed on October 23, 2020).
Rabbi, I. Y., Kayondo, S. I., Bauchet, G., Yusuf, M., Aghogho, C. I., Ogunpaimo, Sedano, J. C. S., Moreno, R. E. M., Mathew, B., Léon, J., Cano, F. A. G., Ballvora,
K., et al. (2020). Genome-wide association analysis reveals new insights into the A., et al. (2017). Major novel QTL for resistance to Cassava bacterial blight
genetic architecture of defensive, agro-morphological and quality-related traits identified through a multi-environmental analysis. Front. Plant Sci. 8:1169.
in cassava. Plant Mol. Biol. 30:20. doi: 10.1007/s11103-020-01038-3 doi: 10.3389/fpls.2017.01169
Rabbi, I. Y., Kulakow, P. A., Manu-Aduening, J. A., Dankyi, A. A., Asibuo, Sedano, J. S., Eduardo, R., Moreno, M., Calle, F., Ernesto, C., and Carrascal, L.
J. Y., Parkes, E. Y., et al. (2015). Tracking crop varieties using genotyping-by- (2017). QTL identification for cassava bacterial blight under natural infection
sequencing markers: A case study using cassava (Manihot esculenta Crantz). conditions. Acta Biol. Colomb. 22, 19–26
BMC Genet. 16:115. doi: 10.1186/s12863-015-0273-1 Singh, A. K., Krishnan, S. G., Vinod, K. K., Ellur, R. K., Bollinedi, H., Bhowmick,
Rabbi, I. Y., Kulembeka, H. P., Masumba, E., Marri, P. R., and Ferguson, M. (2012). P. K., et al. (2019). Precision breeding with genomic tools: A decade long
An EST-derived SNP and SSR genetic linkage map of cassava (Manihot esculenta journey of molecular breeding in rice. Indian J. Genet. 79, 181–191. doi: 10.
Crantz). Theor. Appl. Genet. 125, 329–342. doi: 10.1007/s00122-012-18361834 31742/ijgpb.79s.1.7
Rabbi, I. Y., Udoh, L. I., Wolfe, M., Parkes, E. Y., Gedil, M. A., Dixon, A., Somo, M., Kulembeka, H., Mtunda, K., Mrema, E., Salum, K., Wolfe, M. D., et al.
et al. (2017). Genome-wide association mapping of correlated traits in cassava: (2020). Genomic prediction and quantitative trait locus discovery in a cassava
Dry matter and total carotenoid content. Plant Genome 3:10. doi: 10.3835/ training population constructed from multiple breeding stages. Crop Sci. 60,
plantgenome2016.09.0094 896–913. doi: 10.1002/csc2.20003
Ragot, M., Bonierbale, M., Weltzein, E., and Genetics, N. F. (2018). From Sonnewald, U., Fernie, A. R., Gruissem, W., Schläpfer, P., Anjanappa, R. B.,
market demand to breeding decisions: A framework. Lima: CGIAR Gender and Chang, S. H., et al. (2020). The Cassava Source-Sink project: opportunities
Breedinfg Initiative. and challenges for crop improvement by metabolic engineering. Plant J. 103,
Ramcharan, A., Baranowski, K., McCloskey, P., Ahmed, B., Legg, J., and Hughes, 1655–1665. doi: 10.1111/tpj.14865
D. P. (2017). Deep learning for image-based cassava disease detection. Front. Soto, J. C., Ortiz, J. F., Perlaza-Jiménez, L., Vásquez, A. X., Lopez-Lavalle, L. A. B.,
Plant Sci. 8:1852. doi: 10.3389/fpls.2017.01852 Mathew, B., et al. (2015). A genetic map of cassava (Manihot esculenta Crantz)
Ramcharan, A., McCloskey, P., Baranowski, K., Mbilinyi, N., Mrisho, L., Ndalahwa, with integrated physical mapping of immunity-related genes. BMC Genom.
M., et al. (2019). A mobile-based deep learning model for cassava disease 16:190. doi: 10.1186/s12864-015-13971394
diagnosis. Front. Plant Sci. 10:272. doi: 10.3389/fpls.2019.00272 Spencer, D. S. C., and Ezedinma, C. (2017). Cassava cultivation in sub-Saharan
Ramu, P., Esuma, W., Kawuki, R., Rabbi, I. Y., Egesi, C., Bredeson, J. V., et al. Africa. 123–148.
(2017). Cassava haplotype map highlights fixation of deleterious mutations Sraphet, S., Boonchanawiwat, A., Thanyasiriwat, T., Boonseng, O., Tabata, S.,
during clonal propagation. Nat. Genet. 49, 959–963. doi: 10.1038/ng.3845 Sasamoto, S., et al. (2011). SSR and EST-SSR-based genetic linkage map of
Resende, M. D., Resende, M. F. Jr., Sansaloni, C. P., Petroli, C. D., Missiaggia, A. A., cassava (Manihot esculenta Crantz). Theor. Appl. Genet. 122, 1161–1170. doi:
Aguiar, A. M., et al. (2012). Genomic selection for growth and wood quality 10.1007/s00122-010-1520-5
in Eucalyptus: capturing the missing heritability and accelerating breeding for Sraphet, S., Boonchanawiwat, A., Thanyasiriwat, T., Thaikert, R., Whankaew, S.,
complex traits in forest trees. New Phytol. 194, 116–128. Smith, D. R., et al. (2017). Quantitative trait loci underlying root yield and
Rodríguez-Sandoval, E., Fernández-Quintero, A., Sandoval-Aldana, A., and starch content in an F1 derived cassava population (Manihot esculenta Crantz).
Cuvelier, G. (2008). Effect of processing conditions on the texture of J. Agric. Sci. 155, 569–581. doi: 10.1017/S0021859616000678
reconstituted cassava dough. Brazilian J. Chem. Eng. 25, 713–722. doi: 10.1590/ Stephenson, K., Amthor, R., Mallowa, S., Nungo, R., Maziya-Dixon, B., Gichuki, S.,
S0104-66322008000400008 et al. (2010). Consuming cassava as a staple food places children 2-5 years old at
Rosales-soto, M. U., Ross, C. F., Younce, F., John, K., Mattinson, D. S., Huber, K., risk for inadequate protein intake, an observational study in Kenya and Nigeria.
et al. (2016). Advances in food technology and nutritional sciences physico- Nutr. J. 9:9. doi: 10.1186/1475-2891-9-9
chemical and sensory evaluation cassava (Manihot Esculenta Crantz) Flour. Tappiban, P., Ying, Y., Pang, Y., Sraphet, S., Srisawad, N., Smith, D. R., et al.
Adv. Food Technol. Nutr. Sci. Open J. 2, 9–18. doi: 10.17140/AFTnSOJ-2-126 (2020). Gelatinization, pasting and retrogradation properties and molecular fine
Said, J. I., Knapka, J. A., Song, M., and Zhang, J. (2015). Cotton QTLdb: structure of starches from seven cassava cultivars. Int. J. Biol. Macromol. 150,
a cotton QTL database for QTL analysis, visualization, and comparison 831–838. doi: 10.1016/j.ijbiomac.2020.02.119
between Gossypium hirsutum and G. hirsutum × G. barbadense Teeken, B., Agbona, A., Abolore, B., Olaosebikan, O., Alamu, E., Adesokan,
populations. Mol. Genet. Genom. 290, 1615–1625. doi: 10.1007/s00438-015- M., et al. (2020). Understanding cassava varietal preferences through pairwise
1021-y ranking of gari−eba and fufu prepared by local farmer−processors. New York:
Sakurai, T., Mochida, K., Yoshida, T., Akiyama, K., Ishitani, M., Seki, M., et al. Wiley.
(2013). Genome-wide discovery and information resource development of Tester, M., and Langridge, P. (2010). Breeding technologies to increase crop
DNA polymorphisms in cassava. PLoS One 8:e74056. doi: 10.1371/journal.pone. production in a changing world. Science 327, 818–822. doi: 10.1126/science.
0074056 1183700
Sambasivam, G., and Opiyo, G. D. (2020). A predictive machine learning Teye, E., Asare, A. P., Amoah, R. S., and Tetteh, J. P. (2011). Determination of the
application in agriculture: Cassava disease detection and classification with dry matter content of cassava (Manihot esculenta Crantz) tubers using specific
imbalanced dataset using convolutional neural networks. Egypt. Inform. J. gravity method. APRN J. Agric. Biol. Sci. 6, 23–28.
3:119. doi: 10.1016/j.eij.2020.02.007 Thanyasiriwat, T., Sraphet, S., Whankaew, S., Boonseng, O., Bao, J., Lightfoot,
Sánchez, T., Ceballos, H., Dufour, D., Ortiz, D., Morante, N., Calle, F., et al. (2014). D. A., et al. (2013). Quantitative trait loci and candidate genes associated with
Prediction of carotenoids, cyanide and dry matter contents in fresh cassava starch pasting viscosity characteristics in cassava (Manihot esculenta Crantz).
root using NIRS and Hunter color techniques. Food Chem. 151, 444–451. doi: Plant Biol. 16, 197–207. doi: 10.1111/plb.12022
10.1016/j.foodchem.2013.11.081 Thirathumthavorn, D., and Trisuth, T. (2008). Gelatinization and retrogradation
Sandoval-Aldana, A., and Fernandez, Q. A. (2013). Physicochemical properties of native and hydroxypropylated crosslinked tapioca starches with
characterization of two cassava (Manihot esculenta Crantz) starches and added sucrose and sodium chloride. Int. J. Food Prop. 11, 858–864. doi: 10.1080/
flours. Sci. Agroaliment. 1, 19–25. 10942910701659567
Santantonio, N., Atanda, S. A., Beyene, Y., Varshney, R. K., Olsen, M., Jones, Tivana, L. D., Da Cruz, Francisco, J., Zelder, F., Bergenståhl, B., and Dejmek,
E., et al. (2020). Strategies for effective use of genomic information in crop P. (2014). Straightforward rapid spectrophotometric quantification of total
breeding programs serving Africa and South Asia. Front. Plant Sci. 11:353. cyanogenic glycosides in fresh and processed cassava products. Food Chem. 158,
doi: 10.3389/fpls.2020.00353 20–27. doi: 10.1016/j.foodchem.2014.02.066
SASHA (2019). The sweetpotato breeding community of practice: Breeding in Africa Tonukari, N. J., Tonukari, N. J., Ezedom, T., Enuma, C. C., Sakpa, S. O., Avwioroko,
for Africa (2009-2019). Available online at: https://www.sweetpotatoknowledge. O. J., et al. (2015). White gold: Cassava as an industrial base. Am. J. Plant Sci. 6,
org/wp-content/uploads/2019/12/01_SASHA_BREEDING_The-Sweetpotato- 972–979. doi: 10.4236/ajps.2015.67103
Frontiers in Genetics | www.frontiersin.org 20 January 2021 | Volume 11 | Article 623736
fgene-11-623736 January 18, 2021 Time: 16:34 # 21
Mbanjo et al. Innovations for Improving Cassava Production
Torres, L. G., De Resende, M. D. V., Azevedo, C. F., Fonseca, E., Silva, F., and cassava mosaic disease resistance and prospects for rapid genetic improvement.
De Oliveira, E. J. (2019). Genomic selection for productive traits in biparental Plant Genome 9:2. doi: 10.3835/plantgenome2015.11.0118
cassava breeding populations. PLoS One 14:e0220245. doi: 10.1371/journal. Wossen, T., Girma, G., Abdoulaye, T., Rabbi, I., Olanrewaju, A., Alene, A.,
pone.0220245 et al. (2017). The Cassava Monitoring Survey in Nigeria.IITA. Ibadan:
Tripathi, J. N., Ntui, V. O., Ron, M., Muiruri, S. K., Britt, A., and Tripathi, L. (2019). Nigeria, 66.
CRISPR/Cas9 editing of endogenous banana streak virus in the B genome of Xia, Z., Zou, M., Zhang, S., Feng, B., and Wang, W. (2014). AFSM sequencing
Musa spp. overcomes a major challenge in banana breeding. Commun. Biol. 2:46. approach: A simple and rapid method for genome-wide SNP and methylation
doi: 10.1038/s42003-019-0288-7 site discovery and genetic mapping. Sci. Rep. 4:7300. doi: 10.1038/srep0
Tripathi, L., Ntui, V. O., and Tripathi, J. N. (2020). CRISPR/Cas9-based 7300
genome editing of banana for disease resistance. Curr. Opin. Plant Biol. 56, Xu, J., Hua, K., and Lang, Z. (2019). Genome editing for horticultural crop
118–126. improvement. Hortic. Res. 6:113. doi: 10.1038/s41438-019-0196-5
Turyagyenda, L. F., Kizito, E. B., Ferguson, M. E., Baguma, Y., Harvey, J. W., Yonemaru, J. i., Yamamoto, T., Fukuoka, S., Uga, Y., Hori, K., and Yano, M.
Gibson, P., et al. (2012). Genetic diversity among farmer-preferred cassava (2010). Q-TARO: QTL annotation rice online database. Rice 3, 194Ű203. doi:
landraces in Uganda. Afr. Crop Sci. J. 20, 15–30. 10.1007/s12284-010-9041-z
Udoh, L. I., Gedil, M., Parkes, E. Y., Kulakow, P., Adesoye, A., Nwuba, C., et al. Yonis, B. O., Pino, del Carpio, D., Wolfe, M., Jannink, J. L., Kulakow, P., et al.
(2017). Candidate gene sequencing and validation of SNP markers linked to (2020). Improving root characterisation for genomic prediction in cassava. Sci.
carotenoid content in cassava (Manihot esculenta Crantz). Mol. Breed 37:123. Rep. 10:80003. doi: 10.1038/s41598-020-64963-9
doi: 10.1007/s11032-017-0718-5 Zhang, S., Chen, X., Lu, C., Ye, J., Zou, M., Lu, K., et al. (2018). Genome-wide
Waisundara, V. Y. (2018). Introductory Chapter: Cassava as a Staple Food. doi: association studies of 11 agronomic traits in cassava (Manihot esculenta crantz).
10.5772/intechopen.70324 Front. Plant Sci. 9:503. doi: 10.3389/fpls.2018.00503
Wang, W., Feng, B., Xiao, J., Xia, Z., Zhou, X., Li, P., et al. (2014). Cassava
genome from a wild ancestor to cultivated varieties. Nat. Commun. 5:5110. Conflict of Interest: The authors declare that the research was conducted in the
doi: 10.1038/ncomms6110 absence of any commercial or financial relationships that could be construed as a
Whankaew, S., Poopear, S., Kanjanawattanawong, S., Tangphatsornruang, S., potential conflict of interest.
Boonseng, O., Lightfoot, D. A., et al. (2011). A genome scan for quantitative
trait loci affecting cyanogenic potential of cassava root in an outbred population. Copyright © 2021 Mbanjo, Rabbi, Ferguson, Kayondo, Eng, Tripathi, Kulakow and
BMC Genomics 12:266. doi: 10.1186/1471-2164-12-266 Egesi. This is an open-access article distributed under the terms of the Creative
Wolfe, M. D., Del Carpio, D. P., Alabi, O., Ezenwaka, L. C., Ikeogu, U. N., Kayondo, Commons Attribution License (CC BY). The use, distribution or reproduction in
I. S., et al. (2017). Prospects for genomic selection in cassava breeding. Plant other forums is permitted, provided the original author(s) and the copyright owner(s)
Genome 10:3. doi: 10.3835/plantgenome2017.03.0015 are credited and that the original publication in this journal is cited, in accordance
Wolfe, M. D., Rabbi, I. Y., Egesi, C., Hamblin, M., Kawuki, R., Kulakow, P., et al. with accepted academic practice. No use, distribution or reproduction is permitted
(2016). Genome-wide association and prediction reveals genetic architecture of which does not comply with these terms.
Frontiers in Genetics | www.frontiersin.org 21 January 2021 | Volume 11 | Article 623736