Heat stress is a major limiting factor for crop productivity. Tomato is highly sensitive to heat stress, which can result in a total yield loss. To adapt to current and future heat stress, there is a dire need to develop heat tolerant cultivars. Here, we review recent attempts to improve screening for heat tolerance and to exploit genetic and genomic resources in tomatoes. We provide key factors related to phenotyping environments and traits (morphological, physiological, and metabolic) to be considered to identify and breed thermo-tolerant genotypes. There is significant variability in tomato germplasm that can be harnessed to breed for thermo-tolerance. Based on our review, we propose that the use of advanced backcross populations and chromosome segments substitution lines is the best means to exploit variability for heat tolerance in non-cultivated tomato species. We applied a meta quantitative trait loci (MQTL) analysis on data from four mapping experiments to co-localize QTL associated with heat tolerance traits (e.g., pollen viability, number of pollen, number of flowers, style protrusion, style length). The analysis revealed 13 MQTL of which 11 were composed of a cluster of QTL. Overall, there was a reduction of about 1.5-fold in the confidence interval (CI) of the MQTL (31.82 cM) compared to the average CI of individual QTL (47.4 cM). This confidence interval is still large and additional mapping resolution approaches such as association mapping and multi-parent linkage mapping are needed. Further investigations are required to decipher the genetic architecture of heat tolerance surrogate traits in tomatoes. Genomic selection and new breeding techniques including genome editing and speed breeding hold promise to fast-track development of improved heat tolerance and other farmer- and consumer-preferred traits in tomatoes.

Single Nucleotide Polymorphism (SNP) markers were used in characterization of 113 cowpea accessions comprising of 108 from Ghana and 5 from abroad. Leaf tissues from plants cultivated at the University of Ghana were genotyped at KBioscience in the United Kingdom. Data was generated for 477 SNPs, out of which 458 revealed polymorphism. The results were used to analyze genetic dissimilarity among the accessions using Darwin 5 software. The markers discriminated among all of the cowpea accessions and the dissimilarity values which ranged from 0.006 to 0.63 were used for factorial plot. Unexpected high levels of heterozygosity were observed on some of the accessions. Accessions known to be closely related clustered together in a dendrogram drawn with WPGMA method. A maximum length sub-tree which comprised of 48 core accessions was constructed. The software package structure was used to separate accessions into three groups, and the programme correctly identified varieties that were known hybrids. The hybrids were those accessions with numerous heterozygous loci. The structure plot showed closely related accessions with similar genome patterns. The SNP markers were more efficient in discriminating among the cowpea germplasm than morphological, seed protein polymorphism and simple sequence repeat studies reported earlier on the same collection.

Key message: Historical data from breeding programs can be efficiently used to improve genomic selection accuracy, especially when the training set is optimized to subset individuals most informative of the target testing set. The current strategy for large-scale implementation of genomic selection (GS) at the International Maize and Wheat Improvement Center (CIMMYT) global maize breeding program has been to train models using information from full-sibs in a “test-half-predict-half approach.” Although effective, this approach has limitations, as it requires large full-sib populations and limits the ability to shorten variety testing and breeding cycle times. The primary objective of this study was to identify optimal experimental and training set designs to maximize prediction accuracy of GS in CIMMYT’s maize breeding programs. Training set (TS) design strategies were evaluated to determine the most efficient use of phenotypic data collected on relatives for genomic prediction (GP) using datasets containing 849 (DS1) and 1389 (DS2) DH-lines evaluated as testcrosses in 2017 and 2018, respectively. Our results show there is merit in the use of multiple bi-parental populations as TS when selected using algorithms to maximize relatedness between the training and prediction sets. In a breeding program where relevant past breeding information is not readily available, the phenotyping expenditure can be spread across connected bi-parental populations by phenotyping only a small number of lines from each population. This significantly improves prediction accuracy compared to within-population prediction, especially when the TS for within full-sib prediction is small. Finally, we demonstrate that prediction accuracy in either sparse testing or “test-half-predict-half” can further be improved by optimizing which lines are planted for phenotyping and which lines are to be only genotyped for advancement based on GP.

Dearth of information on extent of genetic variability in cassava limits the genetic improvement of cassava genotypes in Sierra Leone. The aim of this study was to assess the genetic diversity and relationships within 102 cassava genotypes using agro-morphological and single nucleotide polymorphism markers. Morphological classification based on qualitative traits categorized the germplasm into five different groups, whereas the quantitative trait set had four groups. The SNP markers classified the germplasm into three main cluster groups. A total of seven principal components (PCs) in the qualitative and four PCs in the quantitative trait sets accounted for 79.03% and 72.30% of the total genetic variation, respectively. Significant and positive correlations were observed between average yield per plant and harvest index (r = 0.76***), number of storage roots per plant and harvest index (r = 0.33*), height at first branching and harvest index (0.26*), number of storage roots per plant and average yield per plant (r = 0.58*), height at first branching and average yield per plant (r = 0.24*), length of leaf lobe and petiole length (r = 0.38*), number of leaf lobe and petiole length (r = 0.31*), width of leaf lobe and length of leaf lobe (r = 0.36*), number of leaf lobe and length of leaf lobe (r = 0.43*), starch content and dry matter content (r = 0.99***), number of leaf lobe and root dry matter (r = 0.30*), number of leaf lobe and starch content (r = 0.28*), and height at first branching and plant height (r = 0.45**). Findings are useful for conservation, management, short term recommendation for release and genetic improvement of the crop.