Abstract Intercropping dominates smallholder cassava production in sub-Saharan Africa, yet cassava breeding programs evaluate genotypes exclusively under monoculture. Despite consistently reported system-level yield advantages, cassava yield is reduced by 17–51% under intercropping, indicating a need to reduce this competitive disadvantage through breeding. However, the quantitative-genetic foundations of intercrop breeding remain uncharacterized for tropical root-crop systems. We hypothesized that monoculture selection would capture most, but not all, genetic merit for intercropping and that a limited tester set would be sufficient if general mixing ability predominated. We evaluated 120 cassava clones previously selected under monoculture in IITA advanced yield trials, testing them under monoculture and in intercrop with two contrasting cowpea varieties across two years at Ibadan, Nigeria. Spatial mixed models quantified genetic variation, genotype-by-cropping-system interaction, cross-system genetic relationships, realized selection gain, mixing ability, tester effects, and land equivalent ratio. Intercropping reduced cassava fresh root yield by 19%, but total land equivalent ratios exceeded 1.0 for all clones, confirming a system-level land-use advantage. Cassava performance under intercropping was heritable, with estimates of 0.50–0.75, and genotype-by-cropping-system interaction was not significant. Genetic correlations between monoculture and intercrop performance were high and approached unity (rg = 0.92–0.99), and selection efficiency was 26–44% at 10% intensity, confirming that high genetic correlation does not guarantee effective indirect selection. Monoculture selection captured approximately two-thirds of direct intercrop gain. General mixing ability dominated, specific mixing ability was negligible, and producer effects explained 20–47% of intercrop variance. The two architecturally and phenologically contrasting cowpea varieties had limited influence on cassava rankings. Here, we show for the first time that cassava breeding for intercropping can retain monoculture selection during early stages while adding two representative cowpea testers at the advanced-trial stage. This staged strategy aligns cassava breeding with diversified smallholder systems without creating a separate pipeline.

The cassava ( Manihot esculenta) genome has two large introgressions from its wild relative M. glaziovii on chromosomes 1 and 4 that originate from historical hybridization efforts. The 10 Mbp chromosome 1 introgression has been increasing in frequency in African breeding populations due to its statistical association with higher dry matter content and root number. However, the region also exhibits suppressed recombination, hindering breeders’ ability to combine favorable glaziovii alleles with the cultivated esculenta background. Since homozygous introgressed lines are rarely selected for advanced trials, dominance effects have not been well-characterized. To analyze the effects of the introgression with higher resolution, we generated a population of over 5000 seedlings from crosses between heterozygous introgressed parents and screened for recombinants using ten KASP markers tagging glaziovii- specific alleles. An optimized subset of 453 lines was then selected and evaluated over two years for yield and vigor traits. Unlike previous studies, composite interval mapping and mixed linear models showed no significant associations between glaziovii alleles and dry matter content or root number. Small, opposing effects on clonal vigor were observed at different ends of the introgression. The region showed significant segregation distortion and enrichment of putative deleterious alleles. Genome alignment of M. esculenta and M. glaziovii assemblies did not show any major structural variants in the introgression region, suggesting that suppressed recombination is likely driven by sequence-level divergence rather than structural rearrangements. These results indicate that the glaziovii introgression does not directly contribute to dry matter, supporting the need for recombination and purging of the glaziovii introgression to aid cassava improvement. A large chromosome segment from a wild relative of cassava is an important structural aspect in the cassava genome. Since the chromosome segment tends to be inherited as one block, its effects on cassava traits were not well resolved. Through genetic mapping at higher resolution, we identified that the wild segment impacts early vigor and does not appear to impact dry yield, as was previously thought. While there are no major structural differences between the wild and cultivated chromosome segments, their overall divergence seems to suppress the wild chromosome segment from pairing with the cultivated chromosome segment during reproduction. In the apparent absence of any major benefits from the wild segment, removing it from the breeding population may be beneficial. A set of glaziovii allele-specific markers were designed to track the chromosome 1 introgression haplotype. Segregation distortion suggests the presence of recessive deleterious or lethal alleles in the introgression. Increased recombination is needed to purge deleterious alleles enriched in introgression region. The glaziovii introgression was associated with slightly lower vigor rating and stem diameter. The effects of the previously-identified glaziovii DM QTL were not detected in this population.

Plant breeding in the public sector is a multigenerational process that creates new plant varieties intended to meet current and future needs of society. Many public sector plant breeding programs are over a century old, and they continue to curate plant genetic resources that are far older still. While individual breeders serve as temporary leaders of these programs and the plant genetic resources they maintain, it is only their institutions that have the capacity to provide the necessary generational glue, enabling the accrual of long-term value to both breeders and society. Identifying best practices to ensure mutual benefit to both public sector breeders and their institutions is critical to achieving the smooth leadership transitions necessary for the sustainability and long-term impact of public breeding programs. The findings presented here suggest that the successful passing of the torch in such programs depends not only on strategic institutional support but also, critically, on the routine actions and mindset of the breeders entrusted with their leadership.

Modern domestication efforts have occurred in a wide range of species. These efforts have led to different levels of change from the initial wild populations and market success. In this study, we explore different breeding cycle parameters to understand the rate of change in different potential starting points of neo-domestication breeding programs. The design of a program that will bring a new crop to market requires consideration of cost efficiency and resource allocation. More crop diversity on the market creates an opportunity to design different types of food systems that can be tailored toward regional and local food security. Stochastic simulation of the rate of gain and variance is applied to different parameter combinations through the breeding cycle-crossing, evaluation, and selection-to identify population level changes along the continuum of wild to semi-domestic plant species. The simulated breeding schemes differ in phenotypic gain and variance depending on selection strategy and population type, discovering the largest phenotypic gain of oligogenic traits occurring using phenotypic recurrent selection for landrace and orphan populations while choosing genomic selection for wild populations. There were also differences based on selection strategy, with maximum avoidance consistently leading to lower gains but higher additive variance. Overall, when looking to domesticate a new species, our simulations find phenotypic recurrent selection to be the most cost-effective option and lead to the most gain in early generations of selection, with marker technology being most useful once initial gains have plateaued. Large phenotypic gains can be made in early generation selection. Gains do not necessarily imply that the species will become a marketable crop. Novel genotyping and phenotyping technology impacts costs associated with rate of gain and genetic variance. Using genomic/phenomic technologies in neo-domestication depends on the market targets, species, and life history. Modern domestication efforts have led to different levels of change from the wild populations. In this study, we explore different rates of change when exploring different types of species. Breeding program design in terms of cost and resource allocation greatly impacts what will bring a new crop to market. Simulated breeding schemes differed based on selection strategy. Overall, when looking to domesticate a new species, our simulations find phenotypic recurrent selection to be the most cost-effective option and lead to the most gain in early generations of selection, with marker technology being most useful once initial gains have plateaued.

Climate change is a defining challenge of the 21st century, and this decade is a critical time for action to mitigate the worst effects on human populations and ecosystems. Plant science can play an important role in developing crops with enhanced resilience to harsh conditions (e.g. heat, drought, salt stress, flooding, disease outbreaks) and engineering efficient carbon-capturing and carbon-sequestering plants. Here, we present examples of research being conducted in these areas and discuss challenges and open questions as a call to action for the plant science community.

The assessment of cassava clones across multiple environments is often carried out at the uniform yield trial, a late evaluation stage, before variety release. This is to assess the differential response of the varieties across the testing environments, a phenomenon referred to as genotype-by-environment interaction (GEI). This phenomenon is considered a critical challenge confronted by plant breeders in developing crop varieties. This study used the data from variety trials established as randomized complete block design (RCBD) in three replicates across 11 locations in different agro-ecological zones in Nigeria over four cropping seasons (2016-2017, 2017-2018, 2018-2019, and 2019-2020). We evaluated a total of 96 varieties, including five checks, across 48 trials. We exploited the intricate pattern of GEI by fitting variance-covariance structure models on fresh root yield. The goodness-of-fit statistics revealed that the factor analytic model of order 3 (FA3) is the most parsimonious model based on Akaike Information Criterion (AIC). The three-factor loadings from the FA3 model explained, on average across the 27 environments, 53.5% [FA (1)], 14.0% [FA (2)], and 11.5% [FA (3)] of the genetic effect, and altogether accounted for 79.0% of total genetic variability. The association of factor loadings with weather covariates using partial least squares regression (PLSR) revealed that minimum temperature, precipitation and relative humidity are weather conditions influencing the genotypic response across the testing environments in the southern region and maximum temperature, wind speed, and temperature range for those in the northern region of Nigeria. We conclude that the FA3 model identified the common latent factors to dissect and account for complex interaction in multi-environment field trials, and the PLSR is an effective approach for describing GEI variability in the context of multi-environment trials where external environmental covariables are included in modeling.

Genomic selection has been promising in situations where phenotypic assessments are expensive, laborious, and/or inefficient. This work evaluated the efficiency of genomic prediction methods combined with genetic models in clone and parent selection with the goal of increasing fresh root yield, dry root yield, as well as dry matter content in cassava roots. The bias and predictive ability of the combinations of prediction methods Genomic Best Linear Unbiased Prediction (G-BLUP), Bayes B, Bayes Cπ, and Reproducing Kernel Hilbert Spaces with additive and additive-dominant genetic models were estimated. Fresh and dry root yield exhibited predominantly dominant heritability, while dry matter content exhibited predominantly additive heritability. The combination of prediction methods and genetic models did not show significant differences in the predictive ability for dry matter content. On the other hand, the prediction methods with additive-dominant genetic models had significantly higher predictive ability than the additive genetic models for fresh and dry root yield, allowing higher genetic gains in clone selection. However, higher predictive ability for genotypic values did not result in differences in breeding value predictions between additive and additive-dominant genetic models. G-BLUP with the classical additive-dominant genetic model had the best predictive ability and bias estimates for fresh and dry root yield. For dry matter content, the highest predictive ability was obtained by G-BLUP with the additive genetic model. Dry matter content exhibited the highest heritability, predictive ability, and bias estimates compared with other traits. The prediction methods showed similar selection gains with approximately 67% of the phenotypic selection gain. By shortening the breeding cycle time by 40%, genomic selection may overcome phenotypic selection by 10%, 13%, and 18% for fresh root yield, dry root yield, and dry matter content, respectively, with a selection proportion of 15%. The most suitable genetic model for each trait allows for genomic selection optimization in cassava with high selection gains, thereby accelerating the release of new varieties.

Variety advancement decisions for root quality and yield-related traits in cassava are complex due to the variable patterns of genotype-by-environment interactions (GEI). Therefore, studies focused on the dissection of the existing patterns of GEI using linear-bilinear models such as Finlay-Wilkinson (FW), additive main effect and multiplicative interaction (AMMI), and genotype and genotype-by-environment (GGE) interaction models are critical in defining the target population of environments (TPEs) for future testing, selection, and advancement. This study assessed 36 elite cassava clones in 11 locations over three cropping seasons in the cassava breeding program of IITA based in Nigeria to quantify the GEI effects for root quality and yield-related traits. Genetic correlation coefficients and heritability estimates among environments found mostly intermediate to high values indicating high correlations with the major TPE. There was a differential clonal ranking among the environments indicating the existence of GEI as also revealed by the likelihood ratio test (LRT), which further confirmed the statistical model with the heterogeneity of error variances across the environments fit better. For all fitted models, we found the main effects of environment, genotype, and interaction significant for all observed traits except for dry matter content whose GEI sensitivity was marginally significant as found using the FW model. We identified TMS14F1297P0019 and TMEB419 as two topmost stable clones with a sensitivity values of 0.63 and 0.66 respectively using the FW model. However, GGE and AMMI stability value in conjunction with genotype selection index revealed that IITA-TMS-IBA000070 and TMS14F1036P0007 were the top-ranking clones combining both stability and yield performance measures. The AMMI-2 model clustered the testing environments into 6 mega-environments based on winning genotypes for fresh root yield. Alternatively, we identified 3 clusters of testing environments based on genotypic BLUPs derived from the random GEI component.

Modern breeding methods integrate next-generation sequencing and phenomics to identify plants with the best characteristics and greatest genetic merit for use as parents in subsequent breeding cycles to ultimately create improved cultivars able to sustain high adoption rates by farmers. This data-driven approach hinges on strong foundations in data management, quality control, and analytics. Of crucial importance is a central database able to (1) track breeding materials, (2) store experimental evaluations, (3) record phenotypic measurements using consistent ontologies, (4) store genotypic information, and (5) implement algorithms for analysis, prediction, and selection decisions. Because of the complexity of the breeding process, breeding databases also tend to be complex, difficult, and expensive to implement and maintain. Here, we present a breeding database system, Breedbase (https://breedbase.org/, last accessed 4/18/2022). Originally initiated as Cassavabase (https://cassavabase.org/, last accessed 4/18/2022) with the NextGen Cassava project (https://www.nextgencassava.org/, last accessed 4/18/2022), and later developed into a crop-agnostic system, it is presently used by dozens of different crops and projects. The system is web based and is available as open source software. It is available on GitHub (https://github.com/solgenomics/, last accessed 4/18/2022) and packaged in a Docker image for deployment (https://hub.docker.com/u/breedbase, last accessed 4/18/2022). The Breedbase system enables breeding programs to better manage and leverage their data for decision making within a fully integrated digital ecosystem.

Monoculture cropping systems currently dominate temperate agroecosystems. However, intercropping can provide valuable benefits, including greater yield stability, increased total productivity, and resilience in the face of pest and disease outbreaks. Plant breeding efforts in temperate field crops are largely focused on monoculture production, but as intercropping becomes more widespread, there is a need for cultivars adapted to these cropping systems. Cultivar development for intercropping systems requires a systems approach, from the decision to breed for intercropping systems through the final stages of variety testing and release. Design of a breeding scheme should include information about species variation for performance in intercropping, presence of genotype × management interaction, observation of key traits conferring success in intercropping systems, and the specificity of intercropping performance. Together this information can help to identify an optimal selection scheme. Agronomic and ecological knowledge are critical in the design of selection schemes in cropping systems with greater complexity, and interaction with other researchers and key stakeholders inform breeding decisions throughout the process. This review explores the above considerations through three case studies: (1) forage mixtures, (2) perennial groundcover systems (PGC), and (3) soybean-pennycress intercropping. We provide an overview of each cropping system, identify relevant considerations for plant breeding efforts, describe previous breeding focused on the cropping system, examine the extent to which proposed theoretical approaches have been implemented in breeding programs, and identify areas for future development.

Global efforts are underway to develop cassava with enhanced levels of provitamin A carotenoids to sustainably meet increasing demands for food and nutrition where the crop is a major staple. Herein, we tested the effectiveness of genomic selection (GS) for rapid improvement of cassava for total carotenoids content and associated traits. We evaluated 632 clones from Uganda's provitamin A cassava breeding pipeline and 648 West African introductions. At harvest, each clone was assessed for level of total carotenoids, dry matter content, and resistance to cassava brown streak disease (CBSD). All clones were genotyped with diversity array technology and imputed to a set of 23,431 single nucleotide polymorphic markers. We assessed predictive ability of four genomic prediction methods in scenarios of cross-validation, across population prediction, and inclusion of quantitative trait loci markers. Cross-validations produced the highest mean prediction ability for total carotenoids content (0.52) and the lowest for CBSD resistance (0.20), with G-BLUP outperforming other models tested. Across population, predictions showed low ability of Ugandan population to predict the performance of West African clones, with the highest predictive ability recorded for total carotenoids content (0.34) and the lowest for CBSD resistance (0.12) using G-BLUP. By incorporating chromosome 1 markers associated with carotenoids content as independent kernel in the G-BLUP model of a cross-validation scenario, prediction ability slightly improved from 0.52 to 0.58. These results reinforce ongoing efforts aimed at integrating GS into cassava breeding and demonstrate the utility of this tool for rapid genetic improvement.

Diverse crops are both outbred and clonally propagated. Breeders typically use truncation selection of parents and invest significant time, land, and money evaluating the progeny of crosses to find exceptional genotypes. We developed and tested genomic mate selection criteria suitable for organisms of arbitrary homozygosity level where the full-sibling progeny are of direct interest as future parents and/or cultivars. We extended cross variance and covariance variance prediction to include dominance effects and predicted the multivariate selection index genetic variance of crosses based on haplotypes of proposed parents, marker effects, and recombination frequencies. We combined the predicted mean and variance into usefulness criteria for parent and variety development. We present an empirical study of cassava (Manihot esculenta), a staple tropical root crop. We assessed the potential to predict the multivariate genetic distribution (means, variances, and trait covariances) of 462 cassava families in terms of additive and total value using cross-validation. Most variance (89%) and covariance (70%) prediction accuracy estimates were greater than zero. The usefulness of crosses was accurately predicted with good correspondence between the predicted and the actual mean performance of family members breeders selected for advancement as new parents and candidate varieties. We also used a directional dominance model to quantify significant inbreeding depression for most traits. We predicted 47,083 possible crosses of 306 parents and contrasted them to those previously tested to show how mate selection can reveal the new potential within the germplasm. We enable breeders to consider the potential of crosses to produce future parents (progeny with top breeding values) and varieties (progeny with top own performance).

Plant breeding has been central to global increases in crop yields. Breeding deserves praise for helping to establish better food security, but also shares the responsibility of unintended consequences. Much work has been done describing alternative agricultural systems that seek to alleviate these externalities, however, breeding methods and breeding programs have largely not focused on these systems. Here we explore breeding and selection strategies that better align with these more diverse spatial and temporal agricultural systems.

KEY MESSAGE:GWAS identified eight yield-related, peak starch type of waxy and wild-type starch and 21 starch pasting property-related traits (QTLs). Prediction ability of eight GS models resulted in low to high predictability, depending on trait, heritability, and genetic architecture. Cassava is both a food and an industrial crop in Africa, South America, and Asia, but knowledge of the genes that control yield and starch pasting properties remains limited. We carried out a genome-wide association study to clarify the molecular mechanisms underlying these traits and to explore marker-based breeding approaches. We estimated the predictive ability of genomic selection (GS) using parametric, semi-parametric, and nonparametric GS models with a panel of 276 cassava genotypes from Thai Tapioca Development Institute, International Center for Tropical Agriculture, International Institute of Tropical Agriculture, and other breeding programs. The cassava panel was genotyped via genotyping-by-sequencing, and 89,934 single-nucleotide polymorphism (SNP) markers were identified. A total of 31 SNPs associated with yield, starch type, and starch properties traits were detected by the fixed and random model circulating probability unification (FarmCPU), Bayesian-information and linkage-disequilibrium iteratively nested keyway and compressed mixed linear model, respectively. GS models were developed, and forward predictabilities using all the prediction methods resulted in values of - 0.001-0.71 for the four yield-related traits and 0.33-0.82 for the seven starch pasting property traits. This study provides additional insight into the genetic architecture of these important traits for the development of markers that could be used in cassava breeding programs.

Assembly of a training population (TP) is an important component of effective genomic selection-based breeding programs. In this study, we examined the power of diverse germplasm assembled from two cassava (Manihot esculenta Crantz) breeding programs in Tanzania at different breeding stages to predict traits and discover quantitative trait loci (QTL). This is the first genomic selection and genome-wide association study (GWAS) on Tanzanian cassava data. We detected QTL associated with cassava mosaic disease (CMD) resistance on chromosomes 12 and 16; QTL conferring resistance to cassava brown streak disease (CBSD) on chromosomes 9 and 11; and QTL on chromosomes 2, 3, 8, and 10 associated with resistance to CBSD for root necrosis. We detected a QTL on chromosome 4 and two QTL on chromosome 12 conferring dual resistance to CMD and CBSD. The use of clones in the same stage to construct TPs provided higher trait prediction accuracy than TPs with a mixture of clones from multiple breeding stages. Moreover, clones in the early breeding stage provided more reliable trait prediction accuracy and are better candidates for constructing a TP. Although larger TP sizes have been associated with improved accuracy, in this study, adding clones from Kibaha to those from Ukiriguru and vice versa did not improve the prediction accuracy of either population. Including the Ugandan TP in either population did not improve trait prediction accuracy. This study applied genomic prediction to understand the implications of constructing TP from clones at different breeding stages pooled from different locations on trait accuracy.

Understanding the genetic relationships among farmer-preferred cassava (Manihot esculenta Crantz) varieties is indispensable to genetic improvement efforts. In this study, we present a genetic analysis of 547 samples of cassava grown by 192 smallholder farmers, which were sampled at random within four districts in Uganda. We genotyped these samples at 287,952 single nucleotide polymorphisms using genotyping-by-sequencing and co-analyzed them with 349 cassava samples from the national breeding program in Uganda. The samples collected from smallholders consisted of 86 genetically unique varieties, as assessed using a genetic distance-based approach. Of these varieties, most were cultivated in only one district (30 in Kibaale, 19 in Masindi, 14 in Arua, and three in Apac), and only three were cultivated across all districts. The genetic differentiation we observed among farming districts in Uganda (mean fixation index [F-ST] = .003) is similar to divergence observed within other countries. Despite the fact that none of the breeding lines were directly observed in farmer fields, genetic divergence between the populations was low (F-ST = .020). Interestingly, we detected the presence of introgressions from the wild relative M. glaziovii Mull. Arg. on chromosomes 1 and 4, which implies ancestry with cassava breeding lines. Given the apparently similar pool of alleles in the breeding germplasm, it is likely that breeders have the raw genetic material they require to match the farmer-preferred trait combinations necessary for adoption. Our study highlights the importance of understanding the genetic makeup of cassava currently grown by smallholder farmers and relative to that of plant breeding germplasm.

Cassava is cultivated due to its drought tolerance and high carbohydrate-containing storage roots. The lack of uniformity and irregular shape of storage roots poses constraints on harvesting and post-harvest processing. Here, we phenotyped the Genetic gain and offspring (C1) populations from the International Institute of Tropical Agriculture (IITA) breeding program using image analysis of storage root photographs taken in the field. In the genome-wide association analysis (GWAS), we detected for most shape and size-related traits, QTL on chromosomes 1 and 12. In a previous study, we found the QTL on chromosome 12 to be associated with cassava mosaic disease (CMD) resistance. Because the root uniformity is important for breeding, we calculated the standard deviation (SD) of individual root measurements per clone. With SD measurements we identified new significant QTL for Perimeter, Feret and Aspect Ratio on chromosomes 6, 9 and 16. Predictive accuracies of root size and shape image-extracted traits were mostly higher than yield trait prediction accuracies. This study aimed to evaluate the feasibility of the image phenotyping protocol and assess GWAS and genomic prediction for size and shape image-extracted traits. The methodology described and the results are promising and open up the opportunity to apply high-throughput methods in cassava.

More than 40 QTLs associated with 14 stress-related, quality and agro-morphological traits were identified. A catalogue of favourable SNP markers for MAS and a list of candidate genes are provided. Cassava (Manihot esculenta) is one of the most important starchy root crops in the tropics due to its adaptation to marginal environments. Genetic progress in this clonally propagated crop can be accelerated through the discovery of markers and candidate genes that could be used in cassava breeding programs. We carried out a genome-wide association study (GWAS) using a panel of 5130 clones developed at the International Institute of Tropical Agriculture—Nigeria. The population was genotyped at more than 100,000 SNP markers via genotyping-by-sequencing (GBS). Genomic regions underlying genetic variation for 14 traits classified broadly into four categories: biotic stress (cassava mosaic disease and cassava green mite severity); quality (dry matter content and carotenoid content) and plant agronomy (harvest index and plant type) were investigated. We also included several agro-morphological traits related to leaves, stems and roots with high heritability. In total, 41 significant associations were uncovered. While some of the identified loci matched with those previously reported, we present additional association signals for the traits. We provide a catalogue of favourable alleles at the most significant SNP for each trait-locus combination and candidate genes occurring within the GWAS hits. These resources provide a foundation for the development of markers that could be used in cassava breeding programs and candidate genes for functional validation.

Introgression of alleles from wild relatives has often been adaptive in plant breeding. However, the significance of historical hybridization events in modern breeding is often not clear. Cassava (Manihot esculenta) is among the most important staple foods in the world, sustaining hundreds of millions of people in the tropics, especially in sub-Saharan Africa. Widespread genotyping makes cassava a model for clonally propagated root and tuber crops in the developing world, and provides an opportunity to study the modern benefits and consequences of historical introgression. We detected large introgressed Manihot glaziovii genome-segments in a collection of 2742 modern cassava landraces and elite germplasm, the legacy of a 1930s era breeding to combat disease epidemics. African landraces and improved varieties were, on average, 3.8% (max 13.6%) introgressed. Introgressions accounted for a significant (mean 20%, max 56%) portion of the heritability of tested traits. M. glaziovii alleles on the distal 10 Mb of chr. 1 increased dry matter and root number. On chr. 4, introgressions in a 20 Mb region improved harvest index and brown streak disease tolerance. We observed the introgression frequency on chr. 1 double over three cycles of selection, and that later stage trials selectively excluded homozygotes from consideration as varieties. This indicates a heterozygous advantage of introgressions. However, we also found that maintaining large recombination-suppressed introgressions in the heterozygous state allowed the accumulation of deleterious mutations. We conclude that targeted recombination of introgressions would increase the efficiency of cassava breeding by allowing simultaneous fixation of beneficial alleles and purging of genetic load.

Random forests (RF) was used to correlate spectral responses to known wet chemistry carotenoid concentrations including total carotenoid content (TCC), all-trans β-carotene (ATBC), violaxanthin (VIO), lutein (LUT), 15-cis beta-carotene (15CBC), 13-cis beta-carotene (13CBC), alpha-carotene (AC), 9-cis beta-carotene (9CBC), and phytoene (PHY) from laboratory analysis of 173 cassava root samples in Columbia. The cross-validated correlations between the actual and estimated carotenoid values using RF ranged from 0.62 in PHY to 0.97 in ATBC. The developed models were used to evaluate the carotenoids of 594 cassava clones with spectral information collected across three locations in a national breeding program (NRCRI, Umudike), Nigeria. Both populations contained cassava clones characterized as white and yellow. The NRCRI evaluated phenotypes were used to assess the genetic correlations, conduct genome-wide association studies (GWAS), and genomic predictions. Estimates of genetic correlation showed various levels of the relationship among the carotenoids. The associations between TCC and the individual carotenoids were all significant (P < 0.001) with high positive values (r > 0.75, except in LUT and PHY where r < 0.3). The GWAS revealed significant genomic regions on chromosomes 1, 2, 4, 13, 14, and 15 associated with variation in at least one of the carotenoids. One of the identified candidate genes, phytoene synthase (PSY) has been widely reported for variation in TCC in cassava. On average, genomic prediction accuracies from the single-trait genomic best linear unbiased prediction (GBLUP) and RF as well as from a multiple-trait GBLUP model ranged from ∼0.2 in LUT and PHY to 0.52 in TCC. The multiple-trait GBLUP model gave slightly higher accuracies than the single trait GBLUP and RF models. This study is one of the initial attempts in understanding the genetic basis of individual carotenoids and demonstrates the usefulness of NIRS in cassava improvement.