Vol.:(0123456789) Theoretical and Applied Genetics (2025) 138:160 https://doi.org/10.1007/s00122-025-04949-1 ORIGINAL ARTICLE Genetic mapping and validation of QTL for whitefly resistance in cassava (Manihot esculenta Crantz) Adriana Bohorquez‑Chaux1 · Luis Augusto Becerra Lopez‑Lavalle2 · Vianey Barrera‑Enriquez3 · María Isabel Gómez‑Jiménez1 · Camilo E. Sanchez‑Sarria1 · Luis Fernando Delgado1 · Xiaofei Zhang4 · Winnie Gimode1  Received: 10 February 2025 / Accepted: 30 May 2025 © The Author(s) 2025 Abstract Key message  QTL associated with whitefly resistance were identified in a cassava F2 population and KASP markers applicable in selection for the trait were validated. Abstract  Whitefly species pose a major threat to cassava production in tropical regions causing direct plant damage and transmitting viruses that lead to devastating cassava diseases. Aleurotrachelus socialis whitefly is one of the pests that affect cassava in South America. Developing resistant cassava varieties is the most sustainable control strategy for managing whiteflies. This study aimed to map the quantitative trait loci (QTL) associated with resistance to A. socialis and develop molecular markers to facilitate marker-assisted selection. An F2 cassava population (N = 183) was generated by selfing a highly resistant F1 derived from a cross between ECU72 (resistant) and COL2246 (susceptible) landraces. Phenotyping was performed using an efficient glasshouse screening method and high throughput image analysis of infested leaves (Nymph- star). We identified QTL on chromosomes 1, 2, 5, 6, 8, 9, and 14, with a stable and highly significant QTL on chromosome 8 (MeF2WFly8.1), explaining 35.44% of the phenotypic variation. To enable efficient selection, high-throughput KASP markers were developed and validated across diverse genetic backgrounds. Three SNPs displayed the highest association with whitefly resistance, with Chr08_6483145 as the most effective marker for selection in diverse backgrounds. These markers are provided for improving the efficiency of whitefly resistance breeding in the global cassava community. Introduction Whiteflies are a major threat to cassava (Manihot esculenta Crantz) production, not only causing direct plant damage but also transmitting viruses responsible for cassava mosaic disease (CMD) and cassava brown streak disease (CBSD) ​ (Colvin et al. 2004; Maruthi et al. 2005)​. While these pests pose a significant challenge in many cassava growing regions, they are particularly devastating in sub-Saharan Africa, where frequent super-abundant whitefly outbreaks exacerbate the spread ​(Otim-Nape et al. 2001; Alicai et al. 2007​)​. In addition to significant yield losses caused by the spread of cassava diseases, whiteflies, especially in the Neo- tropics, contribute to direct damage by feeding on the leaf phloem, leading to symptoms such as chlorosis and prema- ture leaf drop ​(Bellotti and Arias 2001; Bellotti et al. 2012)​ . They also excrete sticky and sugary substances that coat the foliage and serve as a substrate for sooty mold ​(Bellotti and Arias 2001; Nelson 2008)​, resulting in a reduction in photosynthesis and subsequently in root yield, if the infesta- tion is prolonged. The most damaging whitefly species in Africa is Bemisia tabaci, which includes more than 40 morphologically indis- tinguishable putative whitefly species distributed worldwide (​de Moya et al. 2019; Mugerwa et al. 2021)​. The damage caused by the B. tabaci species continues to increase, with the areas affected with CBSD also rapidly expanding ​ Communicated by Damaris Odeny. * Winnie Gimode w.gimode@cgiar.org 1 Alliance of Bioversity International and the International Center for Tropical Agriculture (CIAT), Km 17, Recta Cali‑Palmira, A. A. 6713 Cali, Colombia 2 International Center for Agricultural Research in the Dry Areas (ICARDA), BP 6202, Rabat, Morocco 3 European Bioinformatics Institute (EMBL-EBI), Welcome Genome Campus, Hinxton, Cambridge CB10 1SD, UK 4 University of California Davis, Davis, CA 95616, USA http://orcid.org/0000-0002-3146-3695 http://crossmark.crossref.org/dialog/?doi=10.1007/s00122-025-04949-1&domain=pdf Theoretical and Applied Genetics (2025) 138:160 160   Page 2 of 22 (MacFadyen et al. 2018)​. In the Americas, eleven white- fly species have been reported affecting cassava ​(Bellotti et al. 1999), with Aleurotrachelus socialis Bondar reported to cause significant yield losses of up to 79% ​(Vargas & Bellotti, 1981; Farias 1994; Bellotti et al. 1999) primarily through direct plant damage. Although the B. tabaci com- plex is well known for transmitting many plant viruses, A. socialis has not generally been considered a major virus vector. However, past research indicated A. socialis could transmit cassava virus X (Angel et al. 1987, 1989). Fur- thermore, it is hypothesized that A. socialis may act as an aerial vector for cassava frog skin disease (CFSD) in South America, recently established to be caused by torradoviruses (Jimenez et al. 2024), but its involvement in CFSD transmis- sion is currently under investigation. The control of whitefly populations in the fields relies largely on integrated pest management (IPM) measures, including cultural practices and chemical controls ​(Bellotti and Arias 2001; Carabalí et al. 2010)​. The best strategy, which is a cheaper and more environmentally sustainable option for management, would be to utilize cassava varieties with host resistance against the pest. Screening of the germplasm collection at the International Center for ​​ Tropical Agriculture (CIAT) has identified donor lines with whitefly resistance ​(Bellotti et al. 1987; Parsa et al. 2015; Bohorquez-Chaux et al. 2023; Atim et al. 2024), with some sources including ECU72 exhibiting broad resistance not only against A. socialis, but also against various species of Bemisia tabaci (Omongo et al. 2012; Atim et al. 2024). Quantitative trait loci (QTL) associated with whitefly resistance have been described in other crops including tomato, soybean, melon, cotton and cabbage ​(Nombela and Muñiz 2010; Boissot et  al. 2010; Xu et  al. 2011; Broekgaarden et al. 2018; Aslam et al. 2023​)​. However, no QTL for whitefly resistance has been previously reported in cassava. High-resolution trait mapping in cassava is a significant and ongoing challenge. This is largely due to cassava's inherent genetic complexity: it is highly heterozygous, predominantly clonally propagated and possesses complex trait inheritance. Nonetheless, metabolomic and transcriptomic studies have suggested potential mechanisms of resistance associated with A. socialis ​(Nye et  al. 2023; Perez-Fons et  al. 2023)​. To pinpoint the underlying genetic loci, genomics studies are crucial, building upon the insights from these omics approaches regarding resistance to A. socialis. Identification of loci linked to whitefly resistance would enable the development of molecular markers thereby increasing the efficiency of trait introgression into elite cassava. The objective of this study was to identify QTL associated with A. socialis whitefly resistance in an F2 cassava population derived from ECU72, and to develop molecular markers linked to QTL to enable marker-assisted selection for the trait. Materials and methods Plant material The AM1588 F2 mapping population used in this study was developed by selfing an F1 produced from a cross between the whitefly-resistant landrace ECU72 (Bellotti and Arias 2001; Parsa et al. 2015; Atim et al. 2024)​, and the whitefly- susceptible landrace COL2246. ECU72 which is mostly male sterile, is from Ecuador and was used as the female parent, while COL2246 is from Colombia. The crossing blocks were established at the International Center for Tropical Agriculture (CIAT), Palmira, Colombia, where hand pollinations were performed in the field. The F1 individual selected for generating the F2 mapping population, CM8996-199, exhibited high levels of resist- ance similar to its resistant parent, ECU72 (Fig. 1) and pro- duced abundant female and male flowers. It was selfed in the 2018–2019 season, resulting in an F2 progeny of 183 individuals. These progeny were used for mapping and iden- tification of loci associated with whitefly resistance. To validate the QTL identified in the AM1588 population, two sets of genetic backgrounds were used. These included other F2 populations derived from the ECU72 background, as well as CIAT’s genomic selection training population (GS), consisting of cassava accessions of various pedigrees. Specifically, three F2 populations with ECU72 genetic background were used: GM12202 (a pseudo-F2 population derived from CM8996-199 × CM8996-758 with 119 progeny) and two smaller F2 populations, AM1620 (70 progeny) and AM1621 (24 progeny), both derived from GM8586 F1 [ECU72 × TMS60444 (susceptible African line)]. Additionally, 673 accessions from various families of the GS population, cohort 2–2021, were also used for QTL validation (Table S1). Mass rearing of Aleurotrachelus socialis (bondar) and phenotyping A methodology for mass rearing A. socialis, developed for glasshouse assays and capable of producing about 7600 whitefly adults per plant was used (Bohorquez-Chaux et al. 2023). Phenotypic evaluations were conducted in a glasshouse over three years: 2020–2021 for F2 populations, and in 2023 for the GS population using choice experiments. The plants were propagated in one of three methodologies, from in vitro (AM1588), stakes (AM1588 and GS) and micro-stakes (F2 and pseudo-F2 validation populations), Theoretical and Applied Genetics (2025) 138:160 Page 3 of 22  160 F2 sexual seeds were planted to obtain mother plants. The mother plants from AM1588 were subsequently used to obtain meristems for in vitro propagation and to produce stakes in field conditions. This allowed us to evaluate potential differences in the phenotyping experiments based on the propagation method. Micro-stakes were obtained from mother plants from the other F2 for QTL validation. The infestation was made using the methodology described by Bohorquez-Chaux et al. (2023). An incomplete randomized block design was used separately for each propagation method and population. For the evaluation of AM1588, respective resistant and susceptible checks [ECU72 as resistant, CM8996-199 F1 which is the parent of the population, as well as an infestation/susceptible check (COL1468)] were included in each batch of plants. Large white mesh cages (18 m length × 3 m width × 3 m height) located in a glasshouse were used for the evaluation. Each week three batches of 100 plants (total = 300 plants) with at least five completely opened leaves were infested at the same time. Since the number of genotypes to be evaluated exceeded the capacity of each cage, the number of plants per genotype in every experimental batch was unequal. For AM1588, four plants per genotype were evaluated for each propagation method in eighteen and thirteen experimental batches for plants from in vitro and stakes, respectively. For validation, the F2 populations (AM1620 F2, AM1621 F2 and GM12202 pseudo-F2) and GS population were evaluated using the same methodology described for AM1588 F2. Thirty-four (34) days after infestation, the two most infested leaves were collected (L1 and L2) and photographs taken in a photo box. Analysis for nymph count (NC) and percentage of the leaf occupied by nymphs (%Area) were made using the Nymphstar automated nymph counting method ​(Bohorquez- Chaux et al. 2023). In total, 12 traits were calculated and used for downstream analyses, including NC and %Area for leaf 1, leaf 2 and BLUPs from both leaves, in the in vitro and stake datasets (2 × 3 × 2). These traits are coded as L1_ NC, L2_NC, BLUPs_NC and L1_%Area, L2_%Area and BLUPs_%Area in subsequent text. Statistical analysis The statistical analyses were performed in SAS 9.4 software (SAS Institute Inc., Cary, NC) for Linux using the PROC GLIMMIX (generalized linear mixed models—GLMM) procedure. Correlation between the stake and in  vitro datasets was assessed using Pearson correlation (r) analysis in the R stats package ​(R Core Team 2023)​. Analysis of variance (ANOVA) were performed to measure the effect of genotypes, leaves (L1 and L2), and propagation methods (stake or in vitro) on the NC and %Area. The statistical model applied for each analysis is defined as follows: where y represents either the NC or the %Area, µ denotes the overall mean effect, G captures the effect of different cassava genotypes, rep accounts for the different experimental batches, L represents the effect of leaf position (i.e., L1 (younger leaf) and L2), and ε is the residual effect. All effects were treated as random. Best Linear Unbiased y = � + � + ��� + � + � Fig. 1   Phenotypes of the two cassava genotypes ECU72 (resistant) and COL2246 (sus- ceptible), and their F1 (resistant) 34 days post infestation with A. socialis. CM8996-199 was selfed for the AM1588 mapping population development Theoretical and Applied Genetics (2025) 138:160 160   Page 4 of 22 Predictions (BLUPs) for each accession were calculated for the different experiments and used for QTL identification. The phenotypic distributions of scores from the NC and %Area, for stakes and in vitro, were tested for deviations from normality with Shapiro–Wilk tests (Shapiro and Wilk 1965). Broad sense heritability (H2) ​(Cullis et al. 2006)​ was calculated using the following equation: where VBLUP is the mean–variance difference of two AM1588 F2 individuals based on BLUPs and σ2 g is the genetic variance of these genotypes. A linear mixed-effects model was fitted to estimate the variance components using the R package lme4 (function lmer) ​(Bates et al. 2015). Genotyping, SNP analysis and linkage map construction Total genomic DNA was extracted from fresh young leaf tissues harvested from the AM1588 F2 family and parents, according to Doyle and Hortorium (1991)​. RAD sequencing ​(Davey and Blaxter 2010)​ was performed on 183 genotypes of the AM1588 F2 family at BGI (China). Data were aligned to the Manihot esculenta v6.1 genome (https://​phyto​zome-​next.​jgi.​doe.​gov/​info/​Mescu​lenta_​ v6_1) using BWA-MEM algorithm ​(Li and Durbin 2010)​. Single nucleotide polymorphisms (SNPs) were called and filtered using GATK ​(Depristo et al. 2011)​ , retaining only biallelic SNPs. To ensure high-quality SNPs, additional filters were applied including no more than 10% missingness, ≥ 5% MAF, quality ≥ 30, as well as a depth ≥ 6, using VCFtools ​(Danecek et al. 2011). Finally, a custom Python script was used to remove non- informative SNPs (e.g., identical loci) and those with high segregation distortion (chi-squared test > 5%). The genetic map was constructed using JoinMap v5 (​Van Ooijen J., 2006)​. Since the AM1588 F2 family does not come from two fully homozygous diploid parents (ECU72 and COL2246), heterozygous SNPs were encoded as hkxhk in a.loc file to detect the linkage phases. The results of the phases were then extracted from each linkage group and encoded in an ABH format following this guideline: Initial encoding HH, HK, KK; final encoding to loci with phase {00} A, H, B; final encoding to loci with phase {11} B, H, A. Identical loci and individuals were excluded and linkage groups determined with the independence LOD criterion. The F2 map was constructed using regression mapping and distance between markers was calculated using the Kosambi mapping function (K​ osambi, 1943)​. �� = � − VBLUP∕2� � g QTL mapping, candidate gene identification and marker development QTL mapping was performed using datasets from stakes (6 variables) and in vitro (6 variables) of the AM1588 F2 population. This was determined for L1_NC, L2_NC, BLUPs_NC, L1_%Area, L2_%Area and BLUPs_%Area (total of 12 datasets/traits). Mapping was performed using composite interval mapping (CIM) ​(Zeng 1994)​ in WinQTLCart 2.5 ​(Wang et al. 2007)​ with threshold values calculated through permutation tests (1,000 permutations, α = 0.05) ​(Churchill and Doerge 1994). CIM analysis was performed with a window size of 10 cM using the standard model (Model 6) with a walk speed of 1 cM and 5 marker cofactors determined by forward and backward regression. Candidate genes within the 2-LOD interval of significant QTL were identified using the Manihot esculenta V6 genome (https://​phyto​zome-​next.​jgi.​doe.​gov/​info/​Mescu​ lenta_​v6_1). As a complement, to identify the most significant SNPs associated with whitefly resistance and to compare these results with QTL mapping, a genome-wide association analysis (GWAS) was also performed on the F2 population using BLUPs for the 12 traits. For the GWAS, a final marker set of 247,117 was used following filtering (missingness ≤ 0.1, maf ≥ 0.05). Four models were used for the GWAS on the GAPIT R package ​(Wang and Zhang 2021)​: the General Linear Model (GLM), Fixed and Random Model Circulating Probability Unification (FarmCPU), Bayesian-Information and Linkage- disequilibrium Iteratively nested Keyway (BLINK) and the Multi‐Locus Mixed Model (MLMM) ​(Price et al. 2006; Segura et al. 2012; Liu et al. 2016; Huang et al. 2019)​. We incorporated three principal components (PCs) derived from the genotype matrix, along with a kinship matrix to ensure model robustness. We applied a Bonferroni threshold using the formula α/n where α is 0.05 and n the number of SNPs, and we also applied a false discovery rate (FDR) correction using the Benjamini–Hochberg method to control false positives in our models (Bonferroni 1936; Benjamini and Hochberg 1995). Kompetitive allele specific PCR (KASP) assays for significant SNPs were developed by Intertek Group plc, Australia. Three markers were selected from the peak region of the consistent QTL identified from linkage mapping, and the other thirty-three SNPs obtained from the GWAS (total of 36 markers) for assay development. The allele frequencies and marker quality of the significant SNPs were calculated using a metric that calculates the false positive rate (FPR) and false negative rate (FNR) ​ (Platten et al. 2019; Mbanjo et al. 2024)​. https://phytozome-next.jgi.doe.gov/info/Mesculenta_v6_1 https://phytozome-next.jgi.doe.gov/info/Mesculenta_v6_1 https://phytozome-next.jgi.doe.gov/info/Mesculenta_v6_1 https://phytozome-next.jgi.doe.gov/info/Mesculenta_v6_1 Theoretical and Applied Genetics (2025) 138:160 Page 5 of 22  160 Marker validation Significant markers were validated on the GM12202 pseudo-F2, AM1620 F2 and AM1621 F2 as well as on the GS training populations described. Leaf punches from the samples were sent to Intertek (Australia) for Low-Density SNP Genotyping (LDSG) using thirty-six KASP markers. These samples included 673 samples from different genetic backgrounds of the GS population as well as a total of 365 samples from ECU72 genetic background, developed for whitefly resistance mapping/validation {AM1588 F2 (152), AM1620 F2 (70), AM1621 F2 (24), GM12202 pseudo-F2 (119)}. For analysis, ten whitefly-resistant and susceptible checks were also included as haplotype information for these SNPs was available from previous sequencing. These checks included: ECU72, COL2246, TMS60444, CM8996- 199, CM8996-758, GM8586-103, GM8586-64, PER368, PER415 and PER608. Resistant (R), intermediate (I) and susceptible (S) categories were determined for the ECU72 genetic background and the genomic selection training population. This was done by performing a K-means analysis based on the BLUPs for NC and %Area using the base R package 'stats' ​(R Core Team 2023)​. For each population, we assessed the performance of the K-means analysis when the number of cluster (k) varied from 2 to 10. A quality measure of clusters was estimated based on the sum of squares in which the between-cluster sum of squares was divided by the total within-cluster sum of squares, multiplied by 100. This score ranges from 0 – 100%, in which a higher value indicates a better capture of the variance of clusters, however, a value close to 100% could indicate an overfitting of clustering. Finally, using an Elbow method, the optimal number of clusters per population was determined. Results Whitefly resistance in the F2 mapping population The phenotypic distribution for whitefly resistance following infestation confirmed the quantitative nature of the trait. The distributions slightly deviated from a normal distribution according to the Shapiro–Wilk test for normality (P = 0.002) (Fig. 2). The summary statistics for the 12 traits are presented in Table 1. A significant (P < 0.0001) positive correlation of r = 0.54 and r = 0.52 was observed between the datasets from stakes and in vitro, for NC and %Area, respectively (Fig. S1). ANOVA revealed significant differences (P < 0.0001) among genotypes, replicates and leaves (Table S2) with a calculated broad-sense heritability (H2) of 0.58 and 0.56 for in vitro BLUPs_NC and BLUPs_%Area, and 0.73 and 0.74 for stake BLUPs_NC and BLUPs_%Area, respectively. Genotyping, SNP analysis and map construction A total of 390,234 SNPs were obtained from the RAD sequencing. After filtering, a genetic map was constructed using 2,017 SNPs on 18 chromosomes, with a total length of 2658.2 cM and an average of 112 markers per chromosome. Chromosome 18 was significantly shorter (66 cM) compared to the other chromosomes, despite relatively high density of markers (101) with shorter inter-locus distance, while chro- mosome 1 was the longest (164.1 cM) (Fig. 3; Table S3). QTL identification For the 12 different traits mapped in the AM1588 F2 popula- tion, QTL were identified in chromosomes 1, 2, 5, 6, 8, 9 and 14 with R2 ranging from 0.35%—35.44% (Fig. 4; Table 2). Among these, the QTL on chromosome 8 was the most con- sistent and with high LOD across all twelve traits (maximum LOD = 14.02), with QTL on chromosomes 2, 5, and 14 also consistent across at least 6 traits. These 4 consistent QTL (2, 5, 8 and 14) also had the highest R2 values. Table 1 rep- resents the various QTL for the different traits with their respective LOD scores. Several minor QTL were identified; therefore, the focus was on the consistently overlapping QTL (2, 5, 8 and 14) across traits. For chromosome 8 that had overlapping QTL (MeF2WFly8.1) across all traits (both from in vitro and stakes), the R2 range was 8.4–20.18% for plants from in vitro and 24.13–35.44% for stakes. For chromosome 2, two separate QTL were obtained only from stakes, and none from in vitro data. MeF2WFly2.1 overlapped for all six traits from stake data (R2 = 8.49–21.8%), while MeF2WFly2.2 (R2 = 6.58–7.71%) was identified in the same position for two traits (leaf 2 and the total BLUPs for nymph count). For chromosome 5 QTL (MeF2WFly5.1) which were also only obtained from stake data, the locus for all 6 traits overlapped. The R2 range was 1.42–15.61%. Although the QTL on chromosome 14 had peaks at different positions, there was overlap among traits (both in vitro and stakes) and the R2 range was 0.35–11.72% for in vitro and 7.98–14.46% for stakes (Table 2). Overall, the loci from the stake data explained higher phenotypic variation. This is consistent with the broad sense heritability obtained in these experiments, where the 2021 (stakes) assessments had higher H2 (0.73 NC; 0.74% Area) compared to the moderate H2 in the 2020 (in vitro) assessments (0.58 NC; 0.56%Area). GWAS analysis and marker validation Thirty-three SNPs were identified as significant across the four different models. Only 3 SNPs were consist- ently significant across all models (Chr08_6483145, Theoretical and Applied Genetics (2025) 138:160 160   Page 6 of 22 Fig. 2   Frequency distribution of AM1588 F2 cassava mapping popu- lation using a stakes and b in vitro for nymph counts and percentage of leaf area affected following whitefly infestation (N = 183). Arrows indicate the checks (R = resistant, S = susceptible, F1 = parent of AM1588 F2 mapping population) Table 1   Summary statistics of phenotypic data associated with the 12 traits L1 = leaf 1, L2 = leaf 2, NC = nymph count Plant material Trait Checks AM1588 F2 CM8996-199 (F1) ECU72 COL1468 Mean Range Stake BLUPs_%Area 1.30 0.51 5.40 2.29 0.10–6.82 Stake BLUPs_NC 743.13 402.30 3219.89 1381.50 61.17–3332.63 Stake L1_%Area 1.22 0.42 6.00 2.41 0.45–6.40 Stake L1_NC 776.50 209.42 3290.15 1436.45 382.61–3176.30 Stake L2_%Area 1.28 0.51 4.93 2.15 0.45–5.80 Stake L2_NC 706.73 301.60 3126.00 1326.31 390.36–3074.73 In vitro BLUPs_%Area 1.04 1.60 6.10 4.05 0.0422–10.39 In vitro BLUPs_NC 613.50 605.60 3010.37 1542.00 2.25–4805.33 In vitro L1_%Area 1.36 1.90 7.17 3.70 0.01–12.2 In vitro L2_%Area 0.69 1.07 5.00 2.25 0.04–8.97 In vitro L1_NC 847.50 578.80 3248.07 1840.02 771.97–3425.26 In vitro L2_NC 378.50 330.85 2510.49 1231.30 469.16–2742.46 Theoretical and Applied Genetics (2025) 138:160 Page 7 of 22  160 Chr08_6512259, and Chr08_6512325), while the rest of the SNPs were identified using only the GLM and/ or FarmCPU models (Fig. S2, Tables 3 and S4). The 3 SNPs were within the chromosome 8 peak region from the QTL mapping results. The most significant marker (Chr08_6483145) had an R2 value of 25% in the mapping population. The 36 markers (33 from GWAS and 3 in the peak region from linkage mapping) converted to KASP assays were tested for applicability in marker-assisted selection, of which 12 did not amplify correctly and there- fore were not informative in the analyses. In the ECU72 background (F2 mapping and validation populations), markers 1–18 (Table 3) indicated additive segregation patterns, with heterozygotes and genotypes carrying the homozygous unfavorable allele, exhibiting intermediate and susceptible resistance levels to whitefly, respectively. Genotypes with homozygous favorable allele were highly resistant. In contrast, for SNP markers 19–24 (Table 3), there was a dominance segregation pattern with no dif- ferences in whitefly resistance levels between genotypes carrying the homozygous favorable allele and heterozy- gotes, while the genotypes with homozygous unfavorable allele were significantly different, exhibiting susceptibility to whitefly. Validation of significant SNPs in the ECU72 genetic background and GS training population For the ECU72 genetic background populations (N = 365; AM1588 F2, AM1620 F2, AM1621 F2, and GM12202 pseudo-F2), we observed a clustering quality of 80.58% when k = 3. Therefore, the 365 genotypes were classified in three clusters: 171 as resistant (R), 142 as intermediate (I), and 52 as susceptible (S). Of the 6 SNPs that showed a dominant segregation pattern, 4 (Chr08_5641063, Chr08_3623233, Chr08_3607706, and Chr08_6618913) had acceptable percentages of FPR and FNR, and the other 2 had very high percentages of either FPR or FNR (Table 2). Of the 18 SNPs that showed an additive segregation pattern, 14 (Chr08_6483145, Chr08_6512259, Chr08_6512325, Chr08_6512307, Chr08_6512329, Chr08_6640496, Chr08_6640586, Chr08_6825255, Chr08_3607807, Chr08_3778097, Chr08_7985587, Chr08_3607718, Chr08_3623304, Chr08_5725346) had acceptable percentages of FPR and FNR, and the other 4 had very high percentages of either FPR or FNR (Table 3). For the GS training population, a 78.80% quality measure was calculated when k = 3. The 673 genotypes were therefore classified into three clusters: 228 as Fig. 3   Genetic map of the AM1588 F2 population, using 2,017 SNP markers. The horizontal bars on each chromosome represent mapped SNPs, and the scale bar to the left indicates the chromosome length in cM. The bottom bar with the color gradient represents the intra-locus density, denoting distance between markers Theoretical and Applied Genetics (2025) 138:160 160   Page 8 of 22 resistant (R), 310 as intermediate (I), and 135 as susceptible (S). Based on the validation across the ECU72 background and GS training population, 3 SNPs (Chr08_6483145, Chr08_6512325, and Chr08_6640496), which displayed an additive segregation pattern were the most promising as markers for selection, with Chr08_6483145 best applicable in other genetic backgrounds (Fig. 5). 69.8% of the geno- types evaluated had at least one copy of the favorable allele (C) for the best marker (Chr08_6483145) and were in the R (resistant) or I (intermediate) category. For the other signifi- cant markers, Chr08_6512325 (C) and Chr08_6640496 (G), 52.6% and 84.8% of the evaluated genotypes had at least one copy of the favorable allele. Candidate gene annotation and functional analysis Potential candidate genes within the various QTL were identified (Table S5). Table 4 summarizes genes within the loci that have been previously described as involved in host resistance against whiteflies or other insects. MeF2WFly8.1 harbors 115 annotated genes. Worth highlighting, are genes encoding functions related to resistance against sucking insects, such as those involved in cuticular wax formation, cell wall modification, lignin biosynthesis, antibiosis and association with insect effectors. These genes include MYB106, lectin receptor kinases (LecRLKs), peroxidase 53, strictosidine synthase- like 2 (SSL2), patatin-like phospholipase and trichome birefringence-related gene among others (Table  4). MeF2WFly14.1 and MeF2WFly14.2 contain 522 and 400 genes, respectively, with key genes including MYB domain proteins and other genes involved in lignin biosynthesis, trichome development, and immune response against phloem-feeding insects, such as aphids and whiteflies (Table 3). The genes in MeF2WFly5.1 are 204 and include brassinosteroid insensitive 1 (BRI1), phytocystatin 2 and MYB103 associated with defense mechanisms against insects. The QTL on chromosome 2, MeF2WFly2.1 and MeF2WFly2.2 contain 426 and 299 genes, respectively. These genes include MYB domain proteins, ABA-deficient 4, terpenoid synthase and HXXXD-type acyl-transferase family protein, which are involved in cuticular wax biosynthesis, trichome development, response to aphids and whiteflies, and secondary metabolism (Table 4). Fig. 4   QTL associated with Aleurotrachelus socialis whitefly resistance in the AM1588 F2 cassava population (N = 183) Theoretical and Applied Genetics (2025) 138:160 Page 9 of 22  160 Ta bl e  2   Q ua nt ita tiv e tra it lo ci (Q TL ) a ss oc ia te d w ith A . s oc ia lis w hi te fly in th e A M 15 88 F 2 c as sa va p op ul at io n an d th e co rr es po nd in g 2- LO D su pp or t i nt er va l f or d iff er en t t ra its Tr ai t Pl an t s ou rc e Q TL n am e C hr Pe ak (c M ) LO D a A dd iti ve b D om in an tc 2- LO D in te rv al (c M )d R ig ht fl an ki ng m ar ke r ( M b) Le ft fla nk in g m ar ke r ( M b) R 2 (% )e L1 _N C St ak e M eF 2W Fl y1 .1 1 5. 01 3. 02 47 3. 00 14 8. 70 0– 8. 3 C 01 _4 53 18 69 C 01 _6 64 84 55 3. 60 L2 _N C St ak e M eF 2W Fl y1 .2 1 14 .4 1 3. 47 34 4. 40 17 7. 28 12 .9 –1 5. 2 C 01 _7 05 12 87 C 01 _7 47 73 64 2. 25 L2 _% A re a In v itr o M eF 2W Fl y1 .3 1 10 5. 31 2. 98 −  0. 56 −  0. 53 10 4. 7– 10 7 C 01 _2 62 37 31 1 C 01 _2 65 84 49 4 1. 02 L1 _% A re a St ak e M eF 2W Fl y2 .1 2 40 .0 1 2. 97 −  0. 45 0. 40 30 –4 7. 8 C 02 _2 39 02 74 C 02 _5 49 62 85 8. 49 L2 _N C St ak e M eF 2W Fl y2 .1 2 38 .0 1 5. 24 −  12 3. 92 48 9. 43 36 .1 –4 5. 5 C 02 _3 11 08 67 C 02 _4 32 75 14 9. 22 B LU Ps _N C St ak e M eF 2W Fl y2 .1 2 39 .0 1 4. 88 −  14 6. 97 48 3. 28 36 .6 –4 8 C 02 _3 13 29 83 C 02 _5 49 62 85 9. 33 B LU Ps _% A re a St ak e M eF 2W Fl y2 .1 2 34 .2 1 3. 49 −  0. 46 0. 33 33 .9 –3 6 C 02 _2 70 99 30 C 02 _3 11 08 67 9. 75 L2 _% A re a St ak e M eF 2W Fl y2 .1 2 34 .2 1 4. 51 −  0. 46 0. 43 30 .9 –4 3. 6 C 02 _2 39 02 74 C 02 _4 16 78 10 12 .4 5 L1 _N C St ak e M eF 2W Fl y2 .1 2 40 .0 1 6. 98 −  35 3. 10 48 7. 97 30 .2 –4 7. 4 C 02 _2 39 02 74 C 02 _5 49 62 85 21 .8 0 B LU Ps _N C St ak e M eF 2W Fl y2 .2 2 68 .6 1 3. 17 −  96 .0 2 39 7. 38 65 .6 –7 6. 2 C 02 _9 73 39 10 C 02 _1 23 21 79 7 6. 58 L2 _N C St ak e M eF 2W Fl y2 .2 2 68 .6 1 3. 4 −  11 3. 20 39 7. 01 65 .5 –7 1. 5 C 02 _9 73 39 10 C 02 _1 23 21 79 7 7. 71 B LU Ps _% A re a St ak e M eF 2W Fl y5 .1 5 13 6. 41 2. 91 −  0. 02 −  0. 69 13 4. 5– 14 0. 2 C 05 _2 56 98 42 1 C 05 _2 65 00 91 4 1. 42 L1 _% A re a St ak e M eF 2W Fl y5 .1 5 13 6. 41 3. 32 0. 01 −  0. 85 13 3. 7– 14 0. 2 C 05 _2 56 79 51 8 C 05 _2 65 00 91 4 1. 91 L2 _N C St ak e M eF 2W Fl y5 .1 5 13 5. 41 4. 74 −  34 .9 7 −  44 9. 19 13 4. 5– 14 5. 3 C 05 _2 56 98 42 1 C 05 _2 69 41 21 2 2. 05 L1 _N C St ak e M eF 2W Fl y5 .1 5 15 0. 81 3. 76 12 2. 62 −  33 5. 38 13 4. 1– 15 5. 9 C 05 _2 56 98 42 1 C 05 _2 82 01 76 9 5. 12 L2 _% A re a St ak e M eF 2W Fl y5 .1 5 14 6. 01 2. 57 0. 38 −  0. 48 13 3. 5– 14 8 C 05 _2 56 79 51 8 C 05 _2 70 76 02 2 10 .5 4 B LU Ps _N C St ak e M eF 2W Fl y5 .1 5 14 7. 01 4. 98 23 4. 19 −  36 5. 54 13 5. 1– 14 8 C 05 _2 58 96 54 6 C 05 _2 70 76 02 2 15 .6 1 L1 _N C In v itr o M eF 2W Fl y6 .1 6 15 2. 31 3. 22 33 5. 78 −  25 9. 01 13 9. 3– 15 4. 3 C 06 _2 57 13 85 4 C 06 _2 71 59 37 1 10 .0 2 L1 _N C In v itr o M eF 2W Fl y8 .1 8 54 .0 1 5. 64 50 5. 94 18 1. 92 48 .4 –6 1. 5 C 08 _5 31 78 24 C 08 _7 99 72 96 8. 40 L2 _N C In v itr o M eF 2W Fl y8 .1 8 54 .0 1 6. 17 45 1. 50 77 .8 2 42 .5 –5 3. 9 C 08 _6 64 04 96 C 08 _8 09 76 64 11 .7 5 B LU Ps _N C In v itr o M eF 2W Fl y8 .1 8 54 .0 1 6. 71 47 5. 13 99 .4 6 42 .5 –6 1. 5 C 08 _5 31 78 24 C 08 _7 99 72 96 12 .1 3 L1 _% A re a In v itr o M eF 2W Fl y8 .1 8 54 .0 1 10 .0 1 1. 70 0. 31 48 .1 –6 1. 5 C 08 _5 31 78 24 C 08 _7 99 72 65 18 .1 7 L2 _% A re a In v itr o M eF 2W Fl y8 .1 8 54 .0 1 10 .1 8 1. 34 0. 14 41 .9 –6 3 C 08 _5 31 78 24 C 08 _7 38 01 99 18 .4 3 B LU Ps _% A re a In v itr o M eF 2W Fl y8 .1 8 54 .0 1 11 .1 2 1. 48 0. 18 53 .5 –6 1. 5 C 08 _6 64 04 96 C 08 _7 38 01 99 20 .1 8 L1 _N C St ak e M eF 2W Fl y8 .1 8 52 .1 1 9. 62 53 1. 94 −  11 3. 48 43 .5 –6 2. 6 C 08 _5 72 53 46 C 08 _7 38 01 99 24 .1 3 L1 _% A re a St ak e M eF 2W Fl y8 .1 8 52 .1 1 10 .6 1. 04 −  0. 19 43 .7 –6 2 C 08 _5 72 53 46 C 08 _7 38 01 99 27 .3 5 B LU Ps _% A re a St ak e M eF 2W Fl y8 .1 8 52 .1 1 12 .1 6 0. 99 −  0. 23 42 .5 –6 0 C 08 _5 25 05 67 C 08 _7 30 75 20 31 .0 6 B LU Ps _N C St ak e M eF 2W Fl y8 .1 8 52 .1 1 13 .0 3 58 8. 37 −  68 .5 1 43 .5 –6 2. 6 C 08 _5 25 05 67 C 08 _7 38 01 99 32 .4 6 L2 _N C St ak e M eF 2W Fl y8 .1 8 52 .1 1 13 .0 1 57 0. 68 −  77 .6 9 53 .3 –6 3. 9 C 08 _5 25 05 67 C 08 _6 61 89 13 33 .0 2 L2 _% A re a St ak e M eF 2W Fl y8 .1 8 52 .1 1 14 .0 2 0. 99 −  0. 22 43 .5 –6 1. 5 C 08 _5 25 05 67 C 08 _8 09 76 06 35 .4 4 L2 _N C In v itr o M eF 2W Fl y8 .2 8 15 2. 41 3. 7 19 4. 73 −  45 7. 65 14 7. 9– 15 3. 7 C 08 _3 14 71 31 1 C 08 _3 24 28 24 4 9. 87 L1 _% A re a St ak e M eF 2W Fl y9 .1 9 92 .9 1 3. 04 -0 .4 9 0. 07 87 .7 –9 4. 5 C 09 _1 59 40 82 9 C 09 _1 87 30 27 0 6. 22 B LU Ps _N C St ak e M eF 2W Fl y1 4. 1 14 13 .0 1 3. 28 16 5. 29 −  31 6. 79 12 .2 –1 5. 7 C 14 _2 69 27 87 C 14 _4 48 81 44 7. 98 L1 _N C St ak e M eF 2W Fl y1 4. 1 14 13 .0 1 3. 55 17 8. 84 −  36 0. 52 12 .2 –1 5. 7 C 14 _2 69 27 87 C 14 _4 48 81 44 8. 33 L2 _% A re a St ak e M eF 2W Fl y1 4. 1 14 13 .0 1 3. 39 0. 32 −  0. 47 11 .5 –2 4. 2 C 14 _2 50 58 16 C 14 _5 11 80 96 8. 50 B LU Ps _% A re a St ak e M eF 2W Fl y1 4. 1 14 13 .0 1 4. 89 0. 42 −  0. 59 12 .2 –1 4. 9 C 14 _2 69 27 87 C 14 _3 06 80 61 12 .8 0 Theoretical and Applied Genetics (2025) 138:160 160   Page 10 of 22 Discussion Phenotyping and genetic control of whitefly resistance The high-throughput and semi-automated screening method for assessing whitefly resistance responses—Nymphstar ​ (Bohorquez-Chaux et al. 2023)​ enabled efficient phenotyping of the populations used in this study. The phenotypic distribution of the evaluated traits were slightly skewed in the AM1588 F2 mapping population, with transgressive segregation observed in the direction of resistance. The correlation between the two datasets of plants derived from in vitro (2020) and those from stakes (2021) was r = 0.54 for nymph counts, and r = 0.52 for the percentage of leaf area occupied by nymphs. Broad sense heritability estimates (0.57–0.75) suggested a significant proportion of whitefly resistance is genetically controlled. The first linkage map of cassava was published by ​Akano et al., (2002​), using RFLPs and SSRs, followed by SNP-based maps ​(​Masumba et al. 2017; Nzuki et al. 2017; Garcia-Oliveira et al. 2020)​. These maps were generated from bi-parental families with contrasting parents for different traits including CMD, CBSD and cassava green mite (CGM). This study represents the first report of a linkage map from a cassava population generated from selfing (CM8996-199 × CM8996- 199 = AM1588 F2, 183 individuals) and the first genetic mapping for resistance to A. socialis whitefly in cassava. Using high throughput phenotyping and genetic mapping, we identified QTL associated with nymph count and percentage of leaf area occupied by nymphs, on chromosomes 1, 2, 5, 6, 8, 9 and 14, in cassava. Whitefly resistance in cassava appears to be quantitatively controlled, as evidenced by multiple QTL identified in this population, indicating that the trait is controlled by several genes. The stable QTL on chromosome 8 (MeF2WFly8.1), which explained up to 35.44% of the phenotypic variance holds the highest potential for introgression into elite cassava lines due to its consistent effect across traits. Interestingly, MeF2WFly8.1 co-localizes with the QTL underlying resistance to CGM in genetically diverse cassava panels (Ezenwaka et al. 2018; Rabbi et al. 2022) and may be a good target for multi- pest resistance breeding in cassava. Among the detected loci, we focused our attention on QTL on chromosomes 2, 5, 8 and 14, as they were stable across multiple traits, and explained relatively high proportion of phenotypic variation. Cassava defense mechanisms against whiteflies Whiteflies and other phloem-feeding insects manipulate plant physiology to their advantage, altering host metab- olism, reducing photosynthetic efficiency, and inducing a  Lo ga rit hm o f o dd s r at io s a t t he p os iti on o f t he p ea k b  A dd iti ve e ffe ct o f Q TL c  D om in an ce e ffe ct o f Q TL d  Th e Q TL in te rv al o n ge ne tic m ap e  Pe rc en t o f p he no ty pi c va ria nc e ex pl ai ne d by th e Q TL L1  =  le af 1 , L 2 =  le af 2 , N C  =  ny m ph c ou nt Ta bl e  2   (c on tin ue d) Tr ai t Pl an t s ou rc e Q TL n am e C hr Pe ak (c M ) LO D a A dd iti ve b D om in an tc 2- LO D in te rv al (c M )d R ig ht fl an ki ng m ar ke r ( M b) Le ft fla nk in g m ar ke r ( M b) R 2 (% )e L1 _% A re a St ak e M eF 2W Fl y1 4. 1 14 13 .0 1 4. 81 0. 48 −  0. 74 1– 23 C 14 _1 02 10 98 C 14 _4 84 45 95 14 .4 6 L2 _N C In v itr o M eF 2W Fl y1 4. 2 14 75 .9 1 2. 79 31 7. 86 −  74 .3 9 74 .2 –7 8. 7 C 14 _1 18 42 04 5 C 14 _9 98 87 85 7. 35 B LU Ps _% A re a In v itr o M eF 2W Fl y1 4. 2 14 72 .4 1 4. 03 0. 87 −  0. 20 54 .4 –7 2. 9 C 14 _8 08 95 54 C 14 _1 14 00 60 1 9. 20 L2 _% A re a In v itr o M eF 2W Fl y1 4. 2 14 75 .9 1 4. 97 0. 93 −  0. 25 63 .9 –8 7. 4 C 14 _8 86 00 12 C 14 _1 48 68 78 9 11 .7 2 L1 _% A re a In v itr o M eF 2W Fl y1 4. 3 14 37 .0 1 4. 21 0. 70 1. 10 35 .1 –3 7. 5 C 14 _6 98 65 71 C 14 _6 98 58 46 0. 35 B LU Ps _N C In v itr o M eF 2W Fl y1 4. 3 14 37 .0 1 2. 91 11 7. 12 42 1. 18 36 .7 –3 7. 5 C 14 _6 98 67 50 C 14 _6 98 58 46 0. 40 Theoretical and Applied Genetics (2025) 138:160 Page 11 of 22  160 Ta bl e  3   F re qu en ci es o f t he 2 4 m ar ke rs o bt ai ne d us in g G W A S an d Q TL m ap pi ng . F PR a nd F N R a re in cl ud ed fo r e ac h m ar ke r, an d se gr eg at io n pa tte rn s a re sh ow n *   a nd * * re pr es en t m ar ke rs th at w er e id en tifi ed u si ng o nl y as so ci at io n or li nk ag e m ap pi ng , r es pe ct iv el y. A ll ot he r m ar ke rs w er e id en tifi ed u si ng G W A S an d w er e al so w ith in th e Q TL in te rv al . sn pM E0 05 72 , s np M E0 05 75 a nd sn pM E0 05 77 a re In te rte k ID s f or th e re sp ec tiv e m ar ke rs . H om  =  ho m oz yg ou s, H et  =  he te ro zy go us N o M ar ke r A pp ro ac h Fa vo ra bl e al le le U nf av or ab le al le le % H om fa vo ra bl e al le le % H et % H om un fa vo ra bl e al le le FP R (% ) FN R (% ) Se gr eg at io n pa tte rn 1 C hr 08 _6 48 31 45 (s np M E0 05 72 ) G W A S an d w ith in M eF 2W Fl y8 .1 C G 24 .5 45 .3 30 .1 18 25 A dd iti ve 2 C hr 08 _6 51 22 59 G W A S an d w ith in M eF 2W Fl y8 .1 T C 22 .5 31 .5 44 .2 14 .7 55 .7 A dd iti ve 3 C hr 08 _6 51 23 25 (s np M E0 05 75 ) G W A S an d w ith in M eF 2W Fl y8 .1 C T 24 .8 27 .8 43 .1 28 36 .5 A dd iti ve 4 C hr 08 _6 51 23 07 G W A S an d w ith in M eF 2W Fl y8 .1 G A 14 22 .6 63 .8 23 .3 38 .4 A dd iti ve 5 C hr 08 _6 51 23 29 G W A S an d w ith in M eF 2W Fl y8 .1 C T 16 .8 23 .6 45 16 40 .3 A dd iti ve 6 C hr 08 _6 64 04 96 (s np M E0 05 77 ) G W A S an d w ith in M eF 2W Fl y8 .1 G A 45 .9 38 .9 15 .3 39 9. 6 A dd iti ve 7 C hr 08 _6 64 05 86 G W A S an d w ith in M eF 2W Fl y8 .1 T G 22 .2 46 .2 31 14 .3 42 .3 A dd iti ve 8 C hr 08 _6 82 52 55 G W A S an d w ith in M eF 2W Fl y8 .1 A G 8. 4 48 .9 37 .2 3. 2 26 .9 A dd iti ve 9 C hr 08 _3 60 78 07 * O nl y G W A S C A 16 .3 42 .8 40 .8 7 32 .7 A dd iti ve 10 C hr 08 _3 77 80 97 * O nl y G W A S C T 17 49 .5 33 .5 17 .4 26 .9 A dd iti ve 11 C hr 08 _7 98 55 87 G W A S an d w ith in M eF 2W Fl y8 .1 G A 24 .4 48 .9 25 .9 15 .9 28 .8 A dd iti ve 12 C hr 08 _4 04 34 69 * O nl y G W A S A T 28 .3 7. 7 17 .1 58 20 .6 A dd iti ve 13 C hr 08 _7 61 77 32 G W A S an d w ith in M eF 2W Fl y8 .1 C T 75 .5 17 .9 6. 5 68 .8 2 A dd iti ve 14 C hr 08 _3 60 77 18 * O nl y G W A S T C 24 .2 47 .6 27 .5 16 .2 23 A dd iti ve 15 C hr 08 _3 62 33 04 * O nl y G W A S T C 31 .3 24 .1 36 .9 17 .6 17 .3 A dd iti ve 16 C hr 08 _5 62 76 30 G W A S an d w ith in M eF 2W Fl y8 .1 G C 71 .7 20 .9 6. 7 86 13 .3 A dd iti ve 17 C hr 08 _5 41 17 80 G W A S an d w ith in M eF 2W Fl y8 .1 G A 8. 5 30 .4 60 .4 2. 6 67 .3 A dd iti ve 18 C hr 08 _5 72 53 46 ** W ith in M eF 2W Fl y8 .1 T A 38 .9 41 .6 19 .1 43 .2 11 .5 A dd iti ve 19 C hr 08 _5 64 10 63 G W A S an d w ith in M eF 2W Fl y8 .1 T A 36 .2 4. 9 54 .4 36 25 D om in an t 20 C hr 08 _3 62 32 33 * O nl y G W A S C T 42 .2 24 32 .5 19 .8 21 .1 D om in an t 21 C hr 08 _3 60 77 06 * O nl y G W A S G C 47 24 .5 25 .6 27 25 D om in an t 22 C hr 08 _4 07 80 44 * O nl y G W A S G T 66 .2 26 .4 6. 2 52 .3 8 D om in an t 23 C hr 08 _2 60 08 88 O nl y G W A S A G 6. 7 27 .1 63 .3 13 .3 67 .3 D om in an t 24 C hr 08 _6 61 89 13 ** W ith in M eF 2W Fl y8 .1 C T 19 29 .5 45 .5 13 .7 38 D om in an t Theoretical and Applied Genetics (2025) 138:160 160   Page 12 of 22 structural modifications in plant tissues ​(Thompson and Goggin 2006)​. In response, cassava has evolved multiple resistance mechanisms, broadly classified into antibiosis, antixenosis, and hormonal regulation, each mediated by specific genetic pathways. Antibiosis One of the key mechanisms identified in this study is antibiosis, whereby plants produce secondary metabo- lites and other compounds that disrupt insect growth and Fig. 5   Boxplots displaying the performance of three KASP markers for whitefly resistance in a) ECU72 background and b) GS training population, for nymph count and percentage leaf area affected. The asterisks indicate levels of statistical significance. *, **, ***, **** significant at p ≤ 0.05, p ≤ 0.01, p ≤ 0.001, p ≤ 0.0001, respectively; ns = non-significant (p > 0.05) Theoretical and Applied Genetics (2025) 138:160 Page 13 of 22  160 Ta bl e  4   P ot en tia l c an di da te g en es a ss oc ia te d w ith w hi te fly re si st an ce in th e id en tifi ed lo ci Pl an t s ou rc e Tr ai t A pp ro ac h SN P C hr G en e ID G en e de sc rip tio n G en e ho m ol og Re fe re nc es B io lo gi ca l f un ct io n, cr op a nd in se ct sp ec ie s St ak e B LU Ps _N C & L1 _N C M eF 2W Fl y2 .1 C hr 02 _5 12 25 09 2 M an es .0 2G 06 86 00 A bs ci si c ac id (A BA )- de fic ie nt 4 A BA 4 G uo e t a l. (2 02 0) , N ye e t a l. (2 02 3) Al eu ro tra ch el us so ci al is , Be m is ia ta ba ci , Ac yr th os ip ho n pi su m , t om at o, ca ss av a St ak e B LU Ps _N C & L2 _N C M eF 2W Fl y2 .2 C hr 02 _1 06 98 62 5 2 M an es .0 2G 14 45 00 H X X X D -ty pe a cy l- tra ns fe ra se fa m ily pr ot ei n BA H D ac yl tra ns fe ra se s X u et  a l. (2 02 3) , N ye e t a l. (2 02 3) A cy la tio n re ac tio ns of se co nd ar y m et ab ol ite s, se ve ra l c ro ps St ak e B LU Ps _N C , L1 _% A re a, L 1_ N C & L 2_ N C M eF 2W Fl y2 .1 C hr 02 _3 55 18 13 2 M an es .0 2G 04 61 00 m yb d om ai n pr ot ei n 30 A TM Y B 30 , M Y B 30 Zh an g et  a l. (2 01 9) , N ye et  a l. (2 02 3) C ut ic ul ar w ax bi os yn th es is , M al us d om es tic a St ak e B LU Ps _N C , L1 _% A re a, L 1_ N C & L 2_ N C M eF 2W Fl y2 .1 C hr 02 _3 19 10 70 2 M an es .0 2G 04 13 00 m yb d om ai n pr ot ei n 5 A TM Y B 5, M Y B 5 Li e t a l. (2 00 9) Tr ic ho m e m or ph og en es is , Ar ab id op si s St ak e B LU Ps _% A re a & L2 _% A re a M eF 2W Fl y2 .1 C hr 02 _3 11 08 67 2 M an es .0 2G 04 01 00 Pe nt at ric op ep tid e re pe at (P PR ) su pe rfa m ily p ro te in Si m on e t a l. (2 02 0) Ph yl lo co lp a sp , P op ul us tr ic ho ca rp a St ak e B LU Ps _N C , L1 _% A re a, L 1_ N C & L 2_ N C M eF 2W Fl y2 .1 C hr 02 _2 53 02 11 2 M an es .0 2G 03 28 00 ste ro l-4 al ph a- m et hy l ox id as e 1– 1 A TS M O 1, A TS M O 1- 1, SM O 1- 1 B eh m er e t a l. (2 01 3) M yz us p er si ca e, N ic ot ia na ta ba cu m , Ph as eo lu s vu lg ar is St ak e B LU Ps _N C , L1 _% A re a, L 1_ N C & L 2_ N C M eF 2W Fl y2 .1 C hr 02 _2 71 85 91 2 M an es .0 2G 03 52 00 Te rp en oi d sy nt ha se su pe rfa m ily p ro te in B on ca n et  a l. (2 02 0) Re sp on se s t o he rb iv or y, se ve ra l cr op s St ak e B LU Ps _N C & L1 _N C M eF 2W Fl y2 .1 C hr 02 _4 40 23 69 2 M an es .0 2G 05 94 00 Th io re do xi n su pe rfa m ily p ro te in Sy ty ki ew ic s e t a l. (2 02 0) Rh op al os ip hu m pa di & M et op ol op hi um di rh od um , m ai ze St ak e B LU Ps _N C , L2 _% A re a & L1 _N C M eF 2W Fl y5 .1 C hr 05 _2 72 23 32 1 5 M an es .0 5G 19 67 00 PH Y TO C Y ST A TI N 2 A tC Y S2 , C Y S2 M ar tin ez e t a l. (2 01 6) H em ip te ra , A ca ri, an d se ve ra l c ro ps St ak e L1 _N C M eF 2W Fl y5 .1 C hr 05 _2 78 83 49 4 5 M an es .0 5G 20 39 00 W R K Y D N A -b in di ng pr ot ei n 2 A TW R K Y 2, W R K Y 2 Ta ng e t a l. (2 02 1) O st ri ni a fu rn ac al is , m ai ze St ak e L1 _N C M eF 2W Fl y5 .1 C hr 05 _2 80 74 54 9 5 M an es .0 5G 20 57 00 m yb d om ai n pr ot ei n 10 3 M Y B 10 3, M Y B 80 , M S1 88 O hm an e t a l. (2 01 3) sy rin gy l l ig ni n bi os yn th es is , Ar ab id op si s Theoretical and Applied Genetics (2025) 138:160 160   Page 14 of 22 Ta bl e  4   (c on tin ue d) Pl an t s ou rc e Tr ai t A pp ro ac h SN P C hr G en e ID G en e de sc rip tio n G en e ho m ol og Re fe re nc es B io lo gi ca l f un ct io n, cr op a nd in se ct sp ec ie s St ak e B LU Ps _N C , L2 _% A re a & L1 _N C M eF 2W Fl y5 .1 C hr 05 _2 65 23 00 4 5 M an es .0 5G 19 22 00 Tr an sd uc in /W D 40 re pe at -li ke su pe rfa m ily p ro te in G ue rr ie ro e t a l. (2 01 5) C el l w al l bi os yn th es is , Ar ab id op si s St ak e L1 _N C M eF 2W Fl y5 .1 C hr 05 _2 78 67 25 6 5 M an es .0 5G 20 38 00 B R A SS IN O ST ER O ID IN SE N SI TI V E 1 (B R I1 ) B R I1 Pr in ce e t a l. (2 01 4) M yz us p er si ca e, Ar ab id op si s St ak e & In  v itr o A ll tra its G W A S C hr 08 _4 04 34 69 8 M an es .0 8G 04 37 00 Pe ro xi da se 5 3- Re la te d Si ng h et  a l. (2 01 3) , N ye et  a l. (2 02 3) El ic ite d by A pi s cr ac ci vo ra & B . ta ba ci , A . s oc ia lis , C ot to n, to m at o, co w pe a, c as sa va St ak e & In  v itr o A ll tra its M eF 2W Fl y8 .1 C hr 08 _5 72 53 46 8 M an es .0 8G 05 59 00 str ic to si di ne sy nt ha se - lik e 2 SS L2 G u et  a l. (2 02 3) M on ot er pe ne s al ka lo id s sy nt he si s, M ai ze St ak e & In  v itr o A ll tra its G W A S C hr 08 _3 77 80 97 8 M an es .0 8G 04 14 00 D iri ge nt P ro te in 20 -R el at ed D IR p ro te in s Pe i e t a l. (2 02 3) Li gn an bi os yn th es is , Ph ry m a le pt os ta ch ya St ak e & In  v itr o A ll tra its G W A S C hr 08 _4 07 80 44 8 M an es .0 8G 04 40 00 Pr ot ei n Tr ic ho m e B ire fr in ge nc e- re la te d TB L Su n et  a l. (2 02 0) M od ify c el l w al l pr op er tie s t hr ou gh O -a ce ty la tio n, Ar ab id op si s St ak e & In  v itr o A ll tra its G W A S C hr 08 _3 77 80 97 8 M an es .0 8G 04 18 00 Pr ot ei n ST IC H EL - Li ke 2 ST IC H EL Ilg en fr itz e t a l. (2 00 3) Tr ic ho m e br an ch nu m be r, Ar ab id op si s St ak e & In  v itr o A ll tra its M eF 2W Fl y8 .1 C hr 08 _5 36 39 07 8 M an es .0 8G 05 32 00 Pl an t i nv er ta se /p ec tin m et hy le ste ra se in hi bi to r s up er fa m ily PM E Si lv a- Sa nz an a et  a l. (2 01 9) M yz us p er si ca e, Ar ab id op si s St ak e & In  v itr o A ll tra its G W A S C hr 08 _6 48 31 45 8 M an es .0 8G 05 84 00 H al oa ci d de ha lo ge na se - lik e hy dr ol as e (H A D ) su pe rfa m ily p ro te in Ya ng e t a l. (2 01 2) Sy nt he si s o f c ut in & su be rin , Ar ab id op si s St ak e & In  v itr o A ll tra its M eF 2W Fl y8 .1 C hr 08 _5 72 53 46 8 M an es .0 8G 05 57 00 tra ns th yr et in -li ke pr ot ei n TT L K yo un g et  a l. (2 00 4) B R I1 su bs tra te , Ar ab id op si s St ak e & In  v itr o A ll tra its G W A S & M eF 2W Fl y8 .1 C hr 08 _6 64 04 96 , 6, 64 0, 58 6 8 M an es .0 8G 05 97 00 ba si c tra ns cr ip tio n fa ct or 3 A TB TF 3, B TF 3 W an g et  a l. (2 01 4) St re ss re sp on se , Ar ab id op si s St ak e & In  v itr o A ll tra its G W A S C hr 08 _6 48 31 45 8 M an es .0 8G 05 85 00 C 2H 2- lik e zi nc fi ng er pr ot ei n zi nc fi ng er p ro te in ZA T3 -li ke B en yó e t a l. (2 02 3) C el l w al l bi og en es is , Ar ab id op si s St ak e & In  v itr o A ll tra its M eF 2W Fl y8 .1 C hr 08 _5 72 53 46 8 M an es .0 8G 05 58 00 ca lm od ul in b in di ng Ya da v et  a l. (2 02 2) Sp od op te ra sp , G ly ci ne m ax Theoretical and Applied Genetics (2025) 138:160 Page 15 of 22  160 Ta bl e  4   (c on tin ue d) Pl an t s ou rc e Tr ai t A pp ro ac h SN P C hr G en e ID G en e de sc rip tio n G en e ho m ol og Re fe re nc es B io lo gi ca l f un ct io n, cr op a nd in se ct sp ec ie s St ak e & In  v itr o A ll tra its G W A S C hr 08 _5 41 17 80 8 M an es .0 8G 05 42 00 IQ -d om ai n 12 IQ D 12 Le vy e t a l. (2 00 5) M yz us p er si ca e, Ar ab id op si s St ak e & In  v itr o A ll tra its G W A S C hr 08 _6 51 22 59 , 6, 51 2, 30 7, 6, 51 2, 32 5, 6, 51 2, 32 9 8 M an es .0 8G 05 89 00 M em br an e tra ffi ck in g V PS 53 fa m ily p ro te in A TV PS 53 , H IT 1, V PS 53 Ro dr ig ue z et  a l. (2 01 7) M yz us p er si ca e, Ar ab id op si s, So la nu m tu be ro su m St ak e & In  v itr o A ll tra its M eF 2W Fl y8 .1 C hr 08 _5 72 53 46 8 M an es .0 8G 05 60 00 Pa ta tin -li ke ph os ph ol ip as e fa m ily pr ot ei n SD P1 Si m iy u et  a l. (2 02 3) Re gu la te s l ig ni n bi os yn th es is , Ar ab id op si s St ak e & In  v itr o A ll tra its G W A S C hr 08 _6 51 22 59 , 6, 51 2, 30 7, 6, 51 2, 32 5, 6, 51 2, 32 9 8 M an es .0 8G 05 88 00 C on ca na va lin A -li ke le ct in p ro te in k in as e fa m ily p ro te in Ta ng e t a l. (2 02 0) ; Sa uv io n et  a l. (2 00 4) Ps yl lid , Ac yr th os ip ho n pi su m , t om at o, pe a St ak e & In  v itr o A ll tra its G W A S C hr 08 _5 62 76 30 , 5, 64 1, 06 3 8 M an es .0 8G 05 50 00 de hy dr at io n- in du ce d pr ot ei n (E R D 15 ) C ID 1, E R D 15 , LS R 1 K ar io la e t a l. (2 00 6) Er w in ia c ar ot ov or a, Ar ab id op si s St ak e & In  v itr o A ll tra its G W A S C hr 08 _6 82 52 55 8 M an es .0 8G 06 03 00 C D PK -r el at ed k in as e 1 A TC B K 3, A TC R K 1, C R K 1 H et te nh au se n et  a l. (2 01 6) ; Sa nt am ar ia e t a l. (2 01 8) Sp od op te ra e xi gu a, Ap hi s g ly ci ne s, G ly ci ne m ax . St ak e & In  v itr o A ll tra its G W A S C hr 08 _6 48 31 45 8 M an es .0 8G 05 80 00 m yb d om ai n pr ot ei n 10 6 M Y B 10 6, M IX TA -li ke R 2R 3- M Y B fa m ily W an g et  a l. (2 02 0) ; O sh im a et  a l. (2 01 3) C ut ic ul ar w ax fo rm at io n, Ar ab id op si s, Eu sto m a gr an di flo ru m St ak e L1 _% A re a M eF 2W Fl y1 4. 1 C hr 14 _1 31 40 46 14 M an es .1 4G 01 41 00 ET H Y LE N E- IN SE N SI TI V E3 - lik e 3 A tE IL 3, A TS LI M , EI L3 , S LI M 1 H e et  a l. (2 01 7) St re ss re sp on se , Ar ab id op si s St ak e L1 _% A re a M eF 2W Fl y1 4. 1 C hr 14 _2 20 05 38 14 M an es .1 4G 02 63 00 m yb d om ai n pr ot ei n 19 A tM Y B 19 , M Y B 19 W an g et  a l. (2 01 7) M ac ro si ph on ie lla sa nb or ni , C hr ys an th em um m or ifo liu m St ak e A ll tra its M eF 2W Fl y1 4. 1 C hr 14 _2 86 37 92 14 M an es .1 4G 03 44 00 m yb d om ai n pr ot ei n 42 A tM Y B 42 , M Y B 42 Zh on g et  a l. (2 00 8) ; G en g et  a l. (2 02 0) Li gn in b io sy nt he si s du rin g se co nd ar y ce ll w al l fo rm at io n, Ar ab id op si s In v itr o B LU Ps _% A re a & L2 _N C M eF 2W Fl y1 4. 2 C hr 14 _9 95 86 63 14 M an es .1 4G 11 52 00 m yb d om ai n pr ot ei n 82 A tM Y B 82 , M Y B 82 Li an g et  a l. (2 01 4) Tr ic ho m e de ve lo pm en t, Ar ab id op si s Theoretical and Applied Genetics (2025) 138:160 160   Page 16 of 22 development, and form barriers that defend the plant from subsequent insect attack (Li et al. 2022). Plant lectins, such as ConA-like lectins, were identified within MeF2WFly8.1 and play an essential role in plant immunity by disrupting the insect digestion and impairing development (Sauvion et al. 2004; Tang et al. 2020). Similarly, peroxidase 53 con- tributes to whitefly resistance by triggering the reactive oxy- gen species (ROS)-mediated localized cell death, creating a barrier against further infestation (Singh et al. 2013; Nye et al. 2023). Some enzyme inhibitors including phytostatin 2, block pest digestive proteases, reducing their ability to extract nutrients (Martinez et al. 2016). These findings rein- force the hypothesis that cassava uses protein-based defenses to reduce the survival of whitefly nymphs and limit popula- tion expansion. At the biochemical level, jasmonic acid (JA) signaling pathway enhances plant defense responses. Patatin-like phospholipase proteins identified in MeF2WFly8.1 have been reported to increase JA accumulation, deterring insect feeding (Canonne et al. 2011; Simiyu et al. 2023). In addition, flavonoids and phenolics regulated by HXXXD- type acyl-transferase proteins (Xu et al. 2023) identified within MeF2WFly2.2, also contribute to plant resistance by disrupting insect physiology. Volatile organic compounds (VOCs) synthesized by terpenoid synthases, identified within MeF2WFly2.1, may contribute to cassava’s ability to deter whiteflies and in field conditions, attract signal predatory insects (Boncan et al. 2020). Furthermore, the presence of strictosidine synthase-like 2 (SSL2) gene in MeF2WFly8.1 suggests that cassava’s resistance mechanisms include monoterpene alkaloid biosynthesis, which provide chemical defense against herbivores and pathogens in other plant species like maize (Gu et al. 2023). Together, these metabolic processes highlight cassava’s ability to actively manipulate its chemical environment to resist whitefly infestations. Antixenosis Beyond biochemical defenses involved in antibiosis, cassava also relies on structural barriers (antixenosis), preventing insect feeding and egg-laying (Bellotti and Arias 2001). These barriers include waxes, cell wall modifications, changes in cutin or suberin, and the formation of trichomes. Lignin biosynthesis is a well- documented resistance mechanism in plants against sap- sucking insects. Metabolomics and transcriptomic studies have shown that cell wall reinforcement and changes in lignin content contribute to resistance against pests such as A. socialis in cassava (Perez-Fons et al. 2019; Nye et al. 2023). Interestingly Ferguson et al. (2023) found evidence for involvement of the lignin pathway in resistance to CBSD. Notably, key genes responsible for lignin biosynthesis were Ta bl e  4   (c on tin ue d) Pl an t s ou rc e Tr ai t A pp ro ac h SN P C hr G en e ID G en e de sc rip tio n G en e ho m ol og Re fe re nc es B io lo gi ca l f un ct io n, cr op a nd in se ct sp ec ie s In v itr o B LU Ps _% A re a & L2 _N C M eF 2W Fl y1 4. 2 C hr 14 _1 14 00 60 1 14 M an es .1 4G 12 83 00 pl an t n at riu re tic pe pt id e A PN P- A Pa ta né e t a l. (2 02 2) Eff ec to r o f Be m is ia ta ba ci , Ar ab id op si s St ak e A ll tra its M eF 2W Fl y1 4. 1 C hr 14 _2 86 37 92 14 M an es .1 4G 03 58 00 R PM 1 in te ra ct in g pr ot ei n 2 R IN 2 Le e et  a l. (2 01 5) Im m un e re sp on se s, Ar ab id op si s St ak e A ll tra its M eF 2W Fl y1 4. 1 C hr 14 _2 86 37 92 14 M an es .1 4G 03 52 00 SK U 5 si m ila r 2 SK S2 C he n et  a l. (2 02 3) Ro ot c el l w al l fo rm at io n. Ar ab id op si s In v itr o L2 _N C & L2 _% A re a M eF 2W Fl y1 4. 2 C hr 14 _1 18 42 02 2 14 M an es .1 4G 13 33 00 Te tra tri co pe pt id e re pe at (T PR )- lik e su pe rfa m ily p ro te in Zh ou e t a l. (2 02 1) D is ea se re si st an ce , to m at o In v itr o L2 _N C & L2 _% A re a M eF 2W Fl y1 4. 2 C hr 14 _9 82 12 51 14 M an es .1 4G 11 44 00 xy lo gl uc an en do tra ns gl yc os yl as e 6 D iv ol e t a l. (2 00 7) ; N ye et  a l. (2 02 3) Al eu ro tra ch el us so ci al is , M yz us pe rs ic ae , C el er y, Ar ab id op si s, ca ss av a L1  =  le af 1 , L 2 =  le af 2 , N C  =  ny m ph c ou nt Theoretical and Applied Genetics (2025) 138:160 Page 17 of 22  160 located within CBSD resistance QTL. This aligns with earlier findings by Amuge et al. (2017), who demonstrated that these lignin biosynthesis genes were upregulated in CBSD-resistant cassava when challenged with the Ugandan cassava brown streak virus. Consequently, genes involved in the lignin pathway hold significant promise as targets for breeding efforts aimed at developing cassava varieties resistant to both CBSD and whiteflies. In this study, MYB domain proteins involved in plant cuticle formation, trichome development and lignin biosynthesis which are crucial for plant defense were also identified in the QTL. Specific examples are MYB106 that regulates cuticular wax formation (Oshima and Mitsuda 2013; Wang et al. 2020), reported in loci associated with CGM severity (Ezenwaka et al. 2018; Rabbi et al. 2022) and MYB103 involved in lignin deposition and secondary cell wall reinforcement (Ohman et al. 2013), which is related to resistance against A. socialis whitefly in cassava (Nye et al. 2023). Notably, our lead SNP for whitefly resistance (Chr08_6483145) lies approximately 74 kb downstream of S8_6409580, the top marker for CGM resistance in Rabbi et al. (2022), suggesting that this region may harbor gene(s) conferring broad-spectrum resistance. Other MYB domains (19, 42 and 82) were identified, associated with lignin biosynthesis, secondary cell wall thickening, trichome development and aphid resistance (Zhong et al. 2008; Liang et al. 2014; Wang et al. 2017; Geng et al. 2020). Additional genes associated with plant defense against insects through cell wall structure modifications are described in Table 3 (Ilgenfritz et al. 2003; Divol et al. 2007; Yang et al. 2012; Behmer et al. 2013; Guerriero et al. 2015; Silva-Sanzana et al. 2019; Sun et al. 2020; Benyó et al. 2023; Nye et al. 2023; Pei et al. 2023). Trichomes play a role in resistance to pests such as whiteflies, though their role varies by plant species. For example, in tomato and Nicotiana, glandular trichomes producing acyl sugars are associated with resistance to the whitefly Bemisia argentifolii (Liedl 1995), while cassava studies show no strong correlation between trichome density and whitefly resistance (Parsa et al. 2015; Pastório et al. 2023). Genes involved in trichome formation were identified in this study, however from observation of our populations, we did not note any correlation between genotypes that had visible trichomes, with high levels of whitefly resistance. Hormonal signaling Hormonal regulation is another critical component of whitefly resistance, orchestrating defense responses at a systemic level through abscisic acid (ABA), brassinosteroids (BRs), and ethylene (ET). ABA-deficient mutants lack sufficient callose deposition, making them more susceptible to aphids (Acyrthosiphon pisum) and whiteflies (Bemisia tabaci) ​(Guo et al. 2020; Nye et al. 2023). Brassinosteroid signaling, mediated by brassinosteroid insensitive 1 (BRI1) and its co-receptor BRI1-associated kinase 1 (BAK1), supports plant growth and pattern-triggered immunity (PTI) restricting herbivore success, as seen in the green peach aphid's (Myzus persicae) interaction with Arabidopsis (Prince et al. 2014). Other hormone signaling genes identified include ethylene-insensitive3 (EIN3) transcription factors which enhance defense gene expression against herbivores through ethylene signaling ​(He et al. 2017)​. This pathway interacts with JA signaling, inducing secondary metabolite production and ROS accumulation. The synergistic regulation of ABA, BRs, and ET provides cassava with a multi-layered resistance system, reinforcing physical barriers, metabolic defenses, and signaling cascades that deter whitefly infestation and feeding. Overall, the QTL identified in the study harbor several key genes that potentially contribute to increased levels of resistance against whiteflies in cassava by enhancing various plant defense mechanisms. Development and validation of KASP markers for breeding KASP markers have been developed and implemented for multiple traits in cassava. These include markers for CMD, dry matter content, total carotenoid content, and hydrogen cyanide concentration ​(Esuma et al. 2022; Rabbi et al. 2022; Kanaabi et al. 2024; Mbanjo et al. 2024)​. These markers have proven invaluable for accelerating genetic gains and facilitating precision breeding through marker- assisted selection (MAS). In this study, we identified and validated KASP markers associated with whitefly (A. socialis) resistance in cassava. The study primarily focused on A. socialis, however, the whitefly resistance source (ECU72) demonstrates resistance not only against A. socialis, but also against various cryptic species of Bemisia tabaci (Omongo et al. 2012; Atim et al. 2024), with likely similar resistance mechanisms. The validation process involved individuals from multiple genetic backgrounds including progeny derived from ECU72 and other genetic backgrounds in the genomic selection training population of the cassava breeding program at CIAT. Three SNPs displayed the highest association with whitefly resistance, with Chr08_6483145 showing the strongest and most consistent association with whitefly resistance, demonstrating high potential for marker- based breeding applications across diverse genetic backgrounds. The utility of these SNPs for marker-assisted selection should be validated in segregating populations from additional genetic backgrounds in other cassava breeding programs, as they show great potential for the genetic enhancement of whitefly resistance in the global cassava community. As more genomic tools and resources Theoretical and Applied Genetics (2025) 138:160 160   Page 18 of 22 are refined for cassava, marker-assisted breeding will significantly improve genetic gains for vital agronomic and quality traits. Conclusion In this study, we employed linkage and association mapping to identify genetic loci conferring whitefly resistance in a cassava F2 population. Among the identified QTL, MeF2WFly8.1 demonstrated stable expression and accounted for up to 35.44% of the observed phenotypic variance. This stability across various genetic backgrounds highlights its significant potential for enhancing whitefly resistance in cassava. Markers applicable in selection were developed and validated, with Chr08_6483145 emerging as the most significant. This marker shows great promise for use in marker-assisted selection (MAS) to improve whitefly resistance in different cassava genetic backgrounds. Supplementary Information  The online version contains supplemen- tary material available at https://​doi.​org/​10.​1007/​s00122-​025-​04949-1. Acknowledgements  We thank Carmen Adriana Bolanos, Janneth Gutierrez, Luz Andrea Gomez, Juan Cuasquer and the CIAT Cassava Genetics, Entomology and Breeding teams for technical assistance. The authors also thank Paul Fraser, Laura Perez-Fons and Linda Walling for their insightful contributions during the planning and execution of this study. Authors contributions  ABC and LABL-L conceived the project idea, designed the research and secured the financial investment for this project. ABC and MIG conducted glasshouse experiments and phenotyping using images. VB, CS, LFD, ABC, and WG were involved in data analysis. XZ designed the field trials for the genomic selection population used for marker validation. ABC and WG wrote the first draft of the manuscript. All authors contributed to and approved the final manuscript. Funding  This research was funded by the Bill and Melinda Gates Foundation (BMGF) through the Natural Resources Institute (NRI), University of Greenwich, UK, under the African Cassava Whitefly Project Phase II (ACWP II) Project Number: INV-010435. We are also grateful for the genotyping support from the CGIAR Breeding Resources Initiative. Data availability  Data is provided in Electronic Supplementary Material. The leaf images used for phenotyping the mapping population are uploaded on CassavaBase (https://​cassa​vabase.​org/). Declarations  Conflict of interest  The authors declare that they have no conflict of interest. Ethics approval  Not applicable. Open Access  This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. References Akano AO, Dixon AGO, Mba C, Barrera E, Fregene M (2002) Genetic mapping of a dominant gene conferring resistance to cassava mosaic disease. Theor Appl Genet 105(4):521–525. https://​doi.​ org/​10.​1007/​s00122-​002-​0891-7 Alicai T, Omongo CA, Maruthi MN, Hillocks RJ, Baguma Y, Kawuki R, Bua A, Otim-Nape GW, Colvin J (2007) Re-emergence of cassava brown streak disease in Uganda. Plant Dis 91(1):24–29. https://​doi.​org/​10.​1094/​PD-​91-​0024 Amuge T, Berger DK, Katari MS, Myburg AA, Goldman SL, Fergu- son ME (2017) A time series transcriptome analysis of cassava (Manihot esculenta Crantz) varieties challenged with Ugandan cassava brown streak virus. Sci Rep 7:9747. https://​doi.​org/​10.​ 1038/​s41598-​017-​09617-z Angel JC, Pineda LB, Nolt B, Velasco AC (1989) Whiteflies (Homop- tera: Aleyrodidae) associated with the transmission of cassava viruses. Fitopatologia Colombiana 13(2):65–71 Angel SJ, Nolt B & Pineda L, B (1987) Estudios sobre la transmisión por “moscas blancas” (homoptera: aleyrodidae) de virus asocia- dos con el “cuero de sapo” en yuca (manihot esculenta crantz). Universidad Nacional de Colombia Sede Palmira Aslam MH, Khan N, Fatima M, Rashid MA, Davies SJ (2023) Strate- gic replacement of soybean meal with local cotton seed meal on growth performance, body composition, and metabolic health status indicators in the major South Asian carp Catla catla for aquaculture. PLoS ONE. https://​doi.​org/​10.​1371/​journ​al.​pone.​ 02962​20 Atim J, Kalyebi A, Bohorquez-Chaux A, Lopez-Lavalle LAB, Omongo CA, Colvin J, Maruthi MN (2024) Identifying cassava genotypes resistant to the african cassava whitefly, Bemisia tabaci (Gen- nadius). Agriculture 14(7):1016. https://​doi.​org/​10.​3390/​agric​ ultur​e1407​1016 Bates D, Mächler M, Bolker BM, Walker SC (2015) Fitting linear mixed-effects models using lme4. J Stat Softw. https://​doi.​org/​ 10.​18637/​jss.​v067.​i01 Behmer ST, Olszewski N, Sebastiani J, Palka S, Sparacino G, Sciarrno E, Grebenok RJ (2013) Plant phloem sterol content: forms, puta- tive functions, and implications for phloem-feeding insects. Front Plant Sci. https://​doi.​org/​10.​3389/​fpls.​2013.​00370 Bellotti AC, Arias B (2001) Host plant resistance to whiteflies with emphasis on cassava as a case study. Crop Prot 20(9):813–823 Bellotti A, Campo BVH, Hyman G (2012) Cassava production and pest management: present and potential threats in a changing environment. In Tropical Plant Biol 5(1):39–72. https://​doi.​org/​ 10.​1007/​s12042-​011-​9091-4 Bellotti AC, Smith L, & Lapointe SL (1999) Recent advances in cas- sava pest management. In Annu. Rev. Entomol (Vol. 44). www.​ annua​lrevi​ews.​org. Bellotti AC (1987) Cassava breeding: a multidisciplinary review: Workshop: Papers. Centro Internacional de Agricultura Tropical https://doi.org/10.1007/s00122-025-04949-1 https://cassavabase.org/ http://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1007/s00122-002-0891-7 https://doi.org/10.1007/s00122-002-0891-7 https://doi.org/10.1094/PD-91-0024 https://doi.org/10.1038/s41598-017-09617-z https://doi.org/10.1038/s41598-017-09617-z https://doi.org/10.1371/journal.pone.0296220 https://doi.org/10.1371/journal.pone.0296220 https://doi.org/10.3390/agriculture14071016 https://doi.org/10.3390/agriculture14071016 https://doi.org/10.18637/jss.v067.i01 https://doi.org/10.18637/jss.v067.i01 https://doi.org/10.3389/fpls.2013.00370 https://doi.org/10.1007/s12042-011-9091-4 https://doi.org/10.1007/s12042-011-9091-4 http://www.annualreviews.org http://www.annualreviews.org Theoretical and Applied Genetics (2025) 138:160 Page 19 of 22  160 Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Stat Soc: Ser B (Methodol) 57(1):289–300. https://​doi.​org/​10.​1111/j.​ 2517-​6161.​1995.​tb020​31.x Benyó D, Bató E, Faragó D, Rigó G, Domonkos I, Labhane N, Zsig- mond L, Prasad M, Nagy I, Szabados L (2023) The zinc finger protein 3 of Arabidopsis thaliana regulates vegetative growth and root hair development. Front Plant Sci. https://​doi.​org/​10.​3389/​ fpls.​2023.​12215​19 Bohorquez-Chaux A, Gómez-Jiménez MI, Leiva-Sandoval LF, Lopez- Lavalle LAB (2023) Nymphstar: an accurate high-throughput quantitative method for whitefly (Aleurotrachelus socialis Bondar) resistance phenotyping in cassava. Plant Phenome J. https://​doi.​org/​10.​1002/​ppj2.​20089 Boissot N, Thomas S, Sauvion N, Marchal C, Pavis C, Dogimont C (2010) Mapping and validation of QTLs for resistance to aphids and whiteflies in melon. Theor Appl Genet 121(1):9–20. https://​ doi.​org/​10.​1007/​s00122-​010-​1287-8 Boncan DAT, Tsang SSK, Li C, Lee IHT, Lam HM, Chan TF, Hui JHL (2020) Terpenes and terpenoids in plants: Interactions with environment and insects. Int J Mol Sci 21(19):7382. https://​doi.​ org/​10.​3390/​ijms2​11973​82 Bonferroni C (1936) Teoria statistica delle classi e calcolo delle proba- bilita. Pubblicazioni Del R Istituto Superiore di Scienze Econom- iche e Commericiali di Firenze 8:3–62. https://​doi.​org/​10.​4135/​ 97814​12961​288.​n455 Broekgaarden C, Pelgrom KTB, Bucher J, Van Dam NM, Grosser K, Pieterse CMJ, Van Kaauwen M, Steenhuis G, Voorrips RE, De Vos M, Vosman B, Worrich A, Van Wees SCM (2018) Combin- ing QTL mapping with transcriptome and metabolome profiling reveals a possible role for ABA signaling in resistance against the cabbage whitefly in cabbage. PLoS ONE. https://​doi.​org/​10.​ 1371/​journ​al.​pone.​02061​03 Canonne J, Froidure-Nicolas S, Rivas S (2011) Phospholipases in action during plant defense signaling. Plant Signal Behav 6(1):13–18. https://​doi.​org/​10.​4161/​psb.6.​1.​14037 Carabalí A, Bellotti AC, Montoya-Lerma J, Fregene M (2010) Resist- ance to the whitefly, Aleurotrachelus socialis, in wild populations of cassava, Manihot tristis. J Insect Sci 10(1):170 Chen C, Zhang Y, Cai J, Qiu Y, Li L, Gao C, Gao Y, Ke M, Wu S, Wei C, Chen J, Xu T, Friml J, Wang J, Li R, Chao D, Zhang B, Chen X, Gao Z (2023) Multi-copper oxidases SKU5 and SKS1 coordinate cell wall formation using apoplastic redox-based reac- tions in roots. Plant Physiol 192(3):2243–2260. https://​doi.​org/​ 10.​1093/​plphys/​kiad2​07 Churchill GA, Doerge RW (1994) Empirical threshold values for quan- titative trait mapping. Genetics 138(3):963–971 Colvin J, Omongo CA, Maruthi MN, Otim-Nape GW, Thresh JM (2004) Dual begomovirus infections and high Bemisia tabaci populations: two factors driving the spread of a cassava mosaic disease pandemic. Plant Pathol 53:577–584. https://​doi.​org/​10.​ 1111/j.​1365-​3059.​2004.​01062.x Cullis BR, Smith AB, Coombes NE (2006) On the design of early gen- eration variety trials with correlated data. J Agric Biol Environ Stat 11(4):381–393. https://​doi.​org/​10.​1198/​10857​1106X​154443 Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, Handsaker RE, Lunter G, Marth GT, Sherry ST, McVean G, Durbin R (2011) The variant call format and VCFtools. Bioin- formatics 27(15):2156–2158. https://​doi.​org/​10.​1093/​bioin​forma​ tics/​btr330 Davey JL, Blaxter MW (2010) RADseq: next-generation population genetics. Brief Funct Genomics 9(5–6):416–423. https://​doi.​org/​ 10.​1093/​bfgp/​elq031 de Moya RS, Brown JK, Sweet AD, Kimberly KK, Paredes-Montero JR, Waterhouse RM, Johnson KP (2019) Nuclear orthologs derived from whole genome sequencing indicate cryptic diversity in the Bemisia tabaci (Insecta: Aleyrodidae) complex of white- flies. Diversity 11(9):151. https://​doi.​org/​10.​3390/​d1109​0151 Depristo MA, Banks E, Poplin R, Garimella KV, Maguire JR, Hartl C, Philippakis AA, Del Angel G, Rivas MA, Hanna M, McKenna A, Fennell TJ, Kernytsky AM, Sivachenko AY, Cibulskis K, Gabriel SB, Altshuler D, Daly MJ (2011) A framework for variation dis- covery and genotyping using next-generation DNA sequencing data. Nat Genet 43(5):491–501. https://​doi.​org/​10.​1038/​ng.​806 Divol F, Vilaine F, Thibivilliers S, Kusiak C, Sauge MH, Dinant S (2007) Involvement of the xyloglucan endotransglycosylase/ hydrolases encoded by celery XTH1 and Arabidopsis XTH33 in the phloem response to aphids. Plant, Cell Environ 30(2):187– 201. https://​doi.​org/​10.​1111/j.​1365-​3040.​2006.​01618.x Doyle J, & Hortorium B (1991) DNA protocols for plants CTAB Total DNA isolation Esuma W, Eyoo O, Gwandu F, Mukasa S, Alicai T, Ozimati A, Nuwamanya E, Rabbi I, Kawuki R (2022) Validation of KASP markers associated with cassava mosaic disease resistance, stor- age root dry matter and provitamin A carotenoid contents in Ugandan cassava germplasm. Front Plant Sci. https://​doi.​org/​ 10.​3389/​fpls.​2022.​10172​75 Ezenwaka L, Del Carpio DP, Jannink JL, Rabbi I, Danquah E, Asante I, Danquah A, Blay E, Egesi C (2018) Genome-wide association study of resistance to cassava green mite pest and related traits in cassava. Crop Sci 58(5):1907–1918. https://​doi.​org/​10.​2135/​ crops​ci2018.​01.​0024 Farias A (1994) Flutuação populacional de Aleurothrixus aepim em mandioca, em São Miguel das Matas, Bahia. Revista Brasileira Da Mandioca 13:119–122 Ferguson ME, Eyles RP, Garcia-Oliveira AL, Kapinga F, Masumba EA, Amuge T, Bredeson JV, Rokhsar DS, Lyons JB, Shah T, Rounsley S, Mkamilo G (2023) Candidate genes for field resist- ance to cassava brown streak disease revealed through the analy- sis of multiple data sources. Front Plant Sci 14:1270963. https://​ doi.​org/​10.​3389/​fpls.​2023.​12709​63 Garcia-Oliveira LKB, Kasele S, Kapinga F, Masumba E, Mkamilo G, Sichalwe C, Bredeson JV, Lyons JB, Hah T, Muranaka S, Katari MS, Ferguson ME (2020) Genetic analysis and qtl mapping for multiple biotic stress resistance in cassava. PLoS ONE. https://​ doi.​org/​10.​1371/​journ​al.​pone.​02366​74 Geng P, Zhang S, Liu J, Zhao C, Wu J, Cao Y, Fu C, Han X, He H, Zhao Q (2020) MYB20, MYB42, MYB43, and MYB85 regulate phenylalanine and lignin biosynthesis during secondary cell wall formation. Plant Physiol 182(3):1272–1283. https://​doi.​org/​10.​ 1104/​PP.​19.​01070 Gu L, Cao Y, Chen X, Wang H, Zhu B, Du X, Sun Y (2023) The genome-wide identification, characterization, and expression analysis of the strictosidine synthase-like family in maize (Zea mays L.). Int J Mol Sci. https://​doi.​org/​10.​3390/​ijms2​41914​733 Guerriero G, Hausman JF, Ezcurra I (2015) Wd40-repeat proteins in plant cell wall formation: current evidence and research pros- pects. Front Plant Sci. https://​doi.​org/​10.​3389/​fpls.​2015.​01112 Guo H, Sun Y, Yan H, Li C, Ge F (2020) O3-induced priming defense associated with the abscisic acid signaling pathway enhances plant resistance to Bemisia tabaci. Front Plant Sci. https://​doi.​ org/​10.​3389/​fpls.​2020.​00093 He X, Jiang J, Wang CQ, Dehesh K (2017) ORA59 and EIN3 interac- tion couples jasmonate-ethylene synergistic action to antagonis- tic salicylic acid regulation of PDF expression. J Integr Plant Biol 59(4):275–287. https://​doi.​org/​10.​1111/​jipb.​12524 Hettenhausen C, Sun G, He Y, Zhuang H, Sun T, Qi J, Wu J (2016) Genome-wide identification of calcium-dependent protein kinases in soybean and analyses of their transcriptional responses to insect herbivory and drought stress. Scientif Rep. https://​doi.​ org/​10.​1038/​srep1​8973 https://doi.org/10.1111/j.2517-6161.1995.tb02031.x https://doi.org/10.1111/j.2517-6161.1995.tb02031.x https://doi.org/10.3389/fpls.2023.1221519 https://doi.org/10.3389/fpls.2023.1221519 https://doi.org/10.1002/ppj2.20089 https://doi.org/10.1007/s00122-010-1287-8 https://doi.org/10.1007/s00122-010-1287-8 https://doi.org/10.3390/ijms21197382 https://doi.org/10.3390/ijms21197382 https://doi.org/10.4135/9781412961288.n455 https://doi.org/10.4135/9781412961288.n455 https://doi.org/10.1371/journal.pone.0206103 https://doi.org/10.1371/journal.pone.0206103 https://doi.org/10.4161/psb.6.1.14037 https://doi.org/10.1093/plphys/kiad207 https://doi.org/10.1093/plphys/kiad207 https://doi.org/10.1111/j.1365-3059.2004.01062.x https://doi.org/10.1111/j.1365-3059.2004.01062.x https://doi.org/10.1198/108571106X154443 https://doi.org/10.1093/bioinformatics/btr330 https://doi.org/10.1093/bioinformatics/btr330 https://doi.org/10.1093/bfgp/elq031 https://doi.org/10.1093/bfgp/elq031 https://doi.org/10.3390/d11090151 https://doi.org/10.1038/ng.806 https://doi.org/10.1111/j.1365-3040.2006.01618.x https://doi.org/10.3389/fpls.2022.1017275 https://doi.org/10.3389/fpls.2022.1017275 https://doi.org/10.2135/cropsci2018.01.0024 https://doi.org/10.2135/cropsci2018.01.0024 https://doi.org/10.3389/fpls.2023.1270963 https://doi.org/10.3389/fpls.2023.1270963 https://doi.org/10.1371/journal.pone.0236674 https://doi.org/10.1371/journal.pone.0236674 https://doi.org/10.1104/PP.19.01070 https://doi.org/10.1104/PP.19.01070 https://doi.org/10.3390/ijms241914733 https://doi.org/10.3389/fpls.2015.01112 https://doi.org/10.3389/fpls.2020.00093 https://doi.org/10.3389/fpls.2020.00093 https://doi.org/10.1111/jipb.12524 https://doi.org/10.1038/srep18973 https://doi.org/10.1038/srep18973 Theoretical and Applied Genetics (2025) 138:160 160   Page 20 of 22 Huang M, Liu X, Zhou Y, Summers RM, Zhang Z (2019) BLINK: a package for the next level of genome-wide association studies with both individuals and markers in the millions. GigaScience. https://​doi.​org/​10.​1093/​gigas​cience/​giy154 Ilgenfritz H, Bouyer D, Schnittger A, Mathur J, Kirik V, Schwab B, Chua NH, Jürgens G, Hülskamp M (2003) The arabidopsis STICHEL gene is a regulator of trichome branch number and encodes a novel protein. Plant Physiol 131(2):643–655. https://​ doi.​org/​10.​1104/​pp.​014209 Jimenez J, Caicedo S, Pardo JM, Gil-Ordoñez A, Alvarez-Quinto R, Mollov D, Cuellar WJ et al (2024) Single torradovirus infections explain the mysterious cassava frogskin disease in the Americas. Sci Rep 14:29648. https://​doi.​org/​10.​1038/​s41598-​024-​81142-2 Kanaabi M, Mukasa SB, Nuwamanya E, Iragaba P, Baguma JK, Nan- yonjo AR, Wagaba H, Muhumuza N, Namakula FB, Wemba- bazi E, Ozimati A, Kayondo IS, Esuma W, Kawuki RS (2024) Validation of KASP markers associated with hydrogen cyanide in fresh cassava roots in uganda cassava germplasm. Agronomy 14(12):2765. https://​doi.​org/​10.​3390/​agron​omy14​122765 Kariola T, Brader G, Helenius E, Li J, Heino P, Palva ET (2006) Early responsive to dehydration 15, a negative regulator of abscisic acid responses in arabidopsis. Plant Physiol 142(4):1559–1573. https://​doi.​org/​10.​1104/​pp.​106.​086223 Kosambi DD (1943) The estimation of map distances from recombina- tion values. Ann Eugen 12(1):172–175. https://​doi.​org/​10.​1111/j.​ 1469-​1809.​1943.​tb023​21.x Kyoung HN, Li J (2004) The arabidopsis transthyretin-like protein is a potential substrate of Brassinosteroid-Insensitive 1. Plant Cell 16(9):2406–2417. https://​doi.​org/​10.​1105/​tpc.​104.​023903 Lee DH, Bourdais G, Yu G, Robatzek S, Coaker G (2015) Phospho- rylation of the plant immune regulator RPM1-INTERACTING PROTEIN4 enhances plant plasma membrane h+-ATPase activ- ity and inhibits flagellin-triggered immune responses in arabi- dopsis. Plant Cell 27(7):2042–2056. https://​doi.​org/​10.​1105/​tpc.​ 114.​132308 Levy M, Wang Q, Kaspi R, Parrella MP, Abel S (2005) Arabidopsis IQD1, a novel calmodulin-binding nuclear protein, stimulates glucosinolate accumulation and plant defense. Plant J 43(1):79– 96. https://​doi.​org/​10.​1111/j.​1365-​313X.​2005.​02435.x Li H, Durbin R (2010) Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics 26(5):589–595. https://​doi.​org/​10.​1093/​bioin​forma​tics/​btp698 Li SF, Milliken ON, Pham H, Seyit R, Napoli R, Preston J, Koltunow AM, Parish RW (2009) The Arabidopsis MYB5 transcription factor regulates mucilage synthesis, seed coat development, and trichome morphogenesis. Plant Cell 21(1):72–89. https://​doi.​org/​ 10.​1105/​tpc.​108.​063503 Li S, Liu S, Zhang Q, Cui M, Zhao M, Li N, Wang S, Wu R, Zhang L, Cao Y, Wang L (2022) The interaction of ABA and ROS in plant growth and stress resistance. Front Plant Sci 13:1050132. https://​ doi.​org/​10.​3389/​fpls.​2022.​10501​32 Liang G, He H, Li Y, Ai Q, Yu D (2014) MYB82 functions in reg- ulation of trichome development in Arabidopsis. J Exp Bot 65(12):3215–3223. https://​doi.​org/​10.​1093/​jxb/​eru179 Liedl B (1995) Acylsugars of wild tomato Lycopersicon penellii alters settling and reduce oviposition of Bemisia argentifolii (Homop- tera: Aleyrodidae). J Econ Entomol 88:742 Liu X, Huang M, Fan B, Buckler ES, Zhang Z (2016) Iterative usage of fixed and random effect models for powerful and efficient genome-wide association studies. PLoS Genet. https://​doi.​org/​ 10.​1371/​journ​al.​pgen.​10057​67 MacFadyen S, Paull C, Boykin LM, De Barro P, Maruthi MN, Otim M, Kalyebi A, Vassão DG, Sseruwagi P, Tay WT, Delatte H, Seguni Z, Colvin J, Omongo CA (2018) Cassava whitefly, Bemi- sia tabaci (Gennadius) (Hemiptera: Aleyrodidae) in East African farming landscapes: a review of the factors determining abun- dance. Bull Entomol Res 108(5):565–582. https://​doi.​org/​10.​ 1017/​S0007​48531​80000​32 Martinez M, Santamaria ME, Diaz-Mendoza M, Arnaiz A, Carrillo L, Ortego F (2016) Phytocystatins: defense proteins against phy- tophagous insects and acari. In Int J Mol Sci 17(10):1747. https://​ doi.​org/​10.​3390/​ijms1​71017​47 Maruthi MN, Hillocks RJ, Mtunda K, Raya MD, Muhanna M, Kiozia H, Rekha AR, Colvin J, Thresh JM (2005) Transmission of Cas- sava brown streak virus by Bemisia tabaci (Gennadius). J Phy- topathol 153(5):307–312. https://​doi.​org/​10.​1111/j.​1439-​0434.​ 2005.​00974.x Masumba EA, Kapinga F, Mkamilo G, Salum K, Kulembeka H, Roun- sley S, Bredeson JV, Lyons JB, Rokhsar DS, Kanju E, Katari MS, Myburg AA, van der Merwe NA, Ferguson ME (2017) QTL associated with resistance to cassava brown streak and cassava mosaic diseases in a bi-parental cross of two Tanzanian farmer varieties, Namikonga and Albert. Theor Appl Genet 130(10):2069–2090. https://​doi.​org/​10.​1007/​s00122-​017-​2943-z Mbanjo EGN, Ogungbesan A, Agbona A, Akpotuzor P, Toyinbo S, Iluebbey P, Rabbi IY, Peteti P, Wages SA, Norton J, Zhang X, Bohórquez-Chaux A, Mushoriwa H, Egesi C, Kulakow P, Parkes E (2024) Validation of SNP markers for diversity analysis, qual- ity control, and trait selection in a biofortified cassava population. Plants. https://​doi.​org/​10.​3390/​plant​s1316​2328 Mugerwa H, Wang HL, Sseruwagi P, Seal S, Colvin J (2021) Whole- genome single nucleotide polymorphism and mating compatibil- ity studies reveal the presence of distinct species in sub-Saharan Africa Bemisia tabaci whiteflies. Insect Sci 28(6):1553–1566. https://​doi.​org/​10.​1111/​1744-​7917.​12881 Nelson S (2008). Sooty Mold. Plant Disease, 52,1–6 http://​www.​ctahr.​ hawaii.​edu/​freep​ubs Nombela G, Muñiz M (2010) Host plant resistance for the management of Bemisia tabaci: a multi-crop survey with emphasis on tomato. Bionomics and Management of a Global Pest, In Bemisia. https://​ doi.​org/​10.​1007/​978-​90-​481-​2460-2_​14 Nye DG, Irigoyen ML, Perez-Fons L, Bohorquez-Chaux A, Hur M, Medina-Yerena D, Lopez-Lavalle LAB, Fraser PD, Walling LL (2023) Integrative transcriptomics reveals association of abscisic acid and lignin pathways with cassava whitefly resistance. BMC Plant Biol. https://​doi.​org/​10.​1186/​s12870-​023-​04607-y Nzuki I, Katari MS, Bredeson JV, Masumba E, Kapinga F, Salum K, Mkamilo GS, Shah T, Lyons JB, Rokhsar DS, Rounsley S, Myburg AA, Ferguson ME (2017) QTL mapping for pest and disease resistance in cassava and coincidence of some QTL with introgression regions derived from Manihot Glaziovii. Front Plant Sci. https://​doi.​org/​10.​3389/​fpls.​2017.​01168 Ohman D, Demedts B, Kumar M, Gerber L, Gorzsas A, Goeminne G, Hedenström M, Ellis B, Boerjan W, Sundberg B (2013) MYB103 is required for FERULATE-5-HYDROXYLASE expression and syringyl lignin biosynthesis in Arabidopsis stems. Plant J 73(1):63–76. https://​doi.​org/​10.​1111/​tpj.​12018 Omongo C, Kawuki R, Bellotti A, Alicai T, Baguma Y, Maruthi M, Bua A, Colvin J (2012) African cassava whitefly, Bemisia tabaci, Resistance in African and South American cassava genotypes. J Integr Agric 11(2):327–336. https://​doi.​org/​10.​1016/​S2095-​ 3119(12)​60017-3 Van Ooijen JW (2006). JoinMap 4, Software for the calculation of genetic linkage maps in experimental populations. Kyazma B.V., Wageningen, Netherlands Oshima Y, Mitsuda N (2013) The MIXTA-like transcription factor MYB16 is a major regulator of cuticle formation in vegetative organs. Plant Signal Behav. https://​doi.​org/​10.​4161/​psb.​26826 Otim-Nape GW, Alicai T, Thresh JM (2001) Changes in the inci- dence and severity of Cassava mosaic virus disease, varietal https://doi.org/10.1093/gigascience/giy154 https://doi.org/10.1104/pp.014209 https://doi.org/10.1104/pp.014209 https://doi.org/10.1038/s41598-024-81142-2 https://doi.org/10.3390/agronomy14122765 https://doi.org/10.1104/pp.106.086223 https://doi.org/10.1111/j.1469-1809.1943.tb02321.x https://doi.org/10.1111/j.1469-1809.1943.tb02321.x https://doi.org/10.1105/tpc.104.023903 https://doi.org/10.1105/tpc.114.132308 https://doi.org/10.1105/tpc.114.132308 https://doi.org/10.1111/j.1365-313X.2005.02435.x https://doi.org/10.1093/bioinformatics/btp698 https://doi.org/10.1105/tpc.108.063503 https://doi.org/10.1105/tpc.108.063503 https://doi.org/10.3389/fpls.2022.1050132 https://doi.org/10.3389/fpls.2022.1050132 https://doi.org/10.1093/jxb/eru179 https://doi.org/10.1371/journal.pgen.1005767 https://doi.org/10.1371/journal.pgen.1005767 https://doi.org/10.1017/S0007485318000032 https://doi.org/10.1017/S0007485318000032 https://doi.org/10.3390/ijms17101747 https://doi.org/10.3390/ijms17101747 https://doi.org/10.1111/j.1439-0434.2005.00974.x https://doi.org/10.1111/j.1439-0434.2005.00974.x https://doi.org/10.1007/s00122-017-2943-z https://doi.org/10.3390/plants13162328 https://doi.org/10.1111/1744-7917.12881 http://www.ctahr.hawaii.edu/freepubs http://www.ctahr.hawaii.edu/freepubs https://doi.org/10.1007/978-90-481-2460-2_14 https://doi.org/10.1007/978-90-481-2460-2_14 https://doi.org/10.1186/s12870-023-04607-y https://doi.org/10.3389/fpls.2017.01168 https://doi.org/10.1111/tpj.12018 https://doi.org/10.1016/S2095-3119(12)60017-3 https://doi.org/10.1016/S2095-3119(12)60017-3 https://doi.org/10.4161/psb.26826 Theoretical and Applied Genetics (2025) 138:160 Page 21 of 22  160 diversity and cassava production in Uganda. Annals Appl Biol 138(3):313–327 Parsa S, Medina C, Rodríguez V (2015) Sources of pest resistance in cassava. Crop Prot 68:79–84. https://​doi.​org/​10.​1016/j.​cropro.​ 2014.​11.​007 Pastório MA, Hoshino AT, Kitzberger CSG et al (2023) The Leaf color and trichome density influence the whitefly infestation in dif- ferent cassava cultivars. InSects. https://​doi.​org/​10.​3390/​insec​ ts140​10004 Patané JSL, Moreira LM, de Melo Teixeira M, Martins J, Setubal JC, Varani AM (2022) New insights into plant natriuretic peptide evolution: from the lysogenic conversion in Xanthomonas to the lateral transfer to the whitefly Bemisia tabaci. Gene. https://​doi.​ org/​10.​1016/j.​gene.​2022.​146326 Pei Y, Cao W, Yu W, Peng C, Xu W, Zuo Y, Wu W, Hu Z (2023) Identification and functional characterization of the dirigent gene family in Phryma leptostachya and the contribution of PlDIR1 in lignan biosynthesis. BMC Plant Biol. https://​doi.​org/​10.​1186/​ s12870-​023-​04297-6 Perez-Fons L, Bohorquez-Chaux A, Irigoyen ML, Garceau DC, Mor- reel K, Boerjan W, Walling LL, Becerra Lopez-Lavalle LA, Fraser PD (2019) A metabolomics characterisation of natural variation in the resistance of cassava to whitefly. BMC Plant Biol. https://​doi.​org/​10.​1186/​s12870-​019-​2107-1 Perez-Fons L, Ovalle TM, Drapal M, Ospina MA, Gkanogiannis A, Bohorquez-Chaux A, Lopez-Lavalle LAB, Fraser PD (2023) Integrated genetic and metabolic characterization of Latin American cassava (Manihot esculenta) germplasm. Plant Physiol 192(4):2672–2686. https://​doi.​org/​10.​1093/​plphys/​kiad2​69 Platten JD, Cobb JN, Zantua RE (2019) Criteria for evaluating molecu- lar markers: comprehensive quality metrics to improve marker- assisted selection. PLoS ONE. https://​doi.​org/​10.​1371/​journ​al.​ pone.​02105​29 Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D (2006) Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet 38(8):904–909. https://​doi.​org/​10.​1038/​ng1847 Prince DC, Drurey C, Zipfel C, Hogenhout SA (2014) The leucine- rich repeat receptor-like kinase BRASSINOSTEROID INSEN- SITIVE1-ASSOCIATED KINASE1 and the Cytochrome P450 PHYTOALEXIN DEFICIENT3 contribute to innate immunity to aphids in arabidopsis. Plant Physiol 164(4):2207–2219. https://​ doi.​org/​10.​1104/​pp.​114.​235598 R Core Team (2023). R Core Team (2023) R: A Language and Envi- ronment for Statistial Computing. R Foundation for Statistical Computing, (4.3.0) Rabbi IY, Kayondo SI, Bauchet G, Yusuf M, Aghogho CI, Ogunpaimo K, Uwugiaren R, Smith IA, Peteti P, Agbona A, Parkes E, Lydia E, Wolfe M, Jannink JL, Egesi C, Kulakow P (2022) Genome- wide association analysis reveals new insights into the genetic architecture of defensive, agro-morphological and quality-related traits in cassava. Plant Mol Biol 109(3):195–213. https://​doi.​org/​ 10.​1007/​s11103-​020-​01038-3 Rodriguez PA, Escudero-Martinez C, Bos JIB (2017) An aphid effec- tor targets trafficking protein VPS52 in a host-specific manner to promote virulence. Plant Physiol 173(3):1892–1903. https://​doi.​ org/​10.​1104/​pp.​16.​01458 Santamaria ME, Arnaiz A, Gonzalez-Melendi P, Martinez M, Diaz I (2018) Plant perception and short-term responses to phytopha- gous insects and mites. Int J Mol Sci 19(5):1356. https://​doi.​org/​ 10.​3390/​ijms1​90513​56 Sauvion N, Charles H, Febvay G, Rahbé Y (2004) Effects of jackbean lectin (ConA) on the feeding behaviour and kinetics of intoxica- tion of the pea aphid. Acyrthosiphon Pisum Entomol Exp Et Applicata 110(1):31–44. https://​doi.​org/​10.​1111/j.​0013-​8703.​ 2004.​00117.x Segura V, Vilhjálmsson BJ, Platt A, Korte A, Seren Ü, Long Q, Nord- borg M (2012) An efficient multi-locus mixed-model approach for genome-wide association studies in structured populations. Nat Genet 44(7):825–830. https://​doi.​org/​10.​1038/​ng.​2314 Shapiro SS, Wilk MB (1965) An analysis of variance test for normality (complete samples). Biometrika 52(3):591–611 Silva-Sanzana C, Celiz-Balboa J, Garzo E, Marcus SE, Parra-Rojas JP, Rojas B, Olmedo P, Rubilar MA, Rios I, Chorbadjian RA, Fereres A, Knox P, Saez-Aguayo S, Blanco-Herrera F (2019) Pectin methylesterases modulate plant homogalacturonan sta- tus in defenses against the aphid myzus persicae. Plant Cell 31(8):1913–1929. https://​doi.​org/​10.​1105/​tpc.​19.​00136 Simiyu DC, Jang JH, Lee OR (2023) A group III patatin-like phospho- lipase gene pPLAIIIδ regulates lignin biosynthesis and influences the rate of seed germination in Arabidopsis thaliana. Front Plant Sci 14:1212979. https://​doi.​org/​10.​3389/​fpls.​2023.​12129​79 Simon SJ, Tschaplinski TJ, LeBoldus M et al (2020) Host plant genetic control of associated fungal and insect species in a Populus hybrid cross. Ecol Evol 10(11):5119–5134. https://​doi.​org/​10.​ 1002/​ece3.​6266 Singh H, Dixit S, Verma PC, Singh PK (2013) Differential peroxidase activities in three different crops upon insect feeding. Plant Sig- nal Behav. https://​doi.​org/​10.​4161/​psb.​25615 Sun A, Yu B, Zhang Q, Peng Y, Yang J, Sun Y, Qin P, Jia T, Smeekens S, Teng S (2020) MYC2-activated trichome Birefringence-like37 acetylates cell walls and enhances her