High Throughput Phenotyping (HTP) Improves Selection Intensity in Groundnut Breeding Pipeline at ICRISAT

Groundnut (Arachis hypogaea L., 2n = 4x = 40) is mainly grown in semi-arid agro-ecologies of Africa and Asia where a combination of abiotic and biotic stresses compromises its yield and quality. Optimizing the breeding scheme for selection intensity enables breeding programs to realise higher rate of genetic gain for targeted traits as well as optimizes time and resources. Selection intensity can be increased by increasing the number of ‘selection candidates’ (that decreases the proportion (p) of candidates selected). High throughput phenotyping (HTP) and genomic tools can be used to exercise selection on many ‘selection candidates.’ At ICRISAT, a two-step selection approach is being deployed to improve drought tolerance that involves a HTP platform, LeasyScan to select early vigour followed by screening in a Managed Stress Environment (MSE). The digital biomass, leaf area index and plant height are used as selection criteria for early vigour. Accordingly, the populations of groundnut breeding pipeline were advanced from F2 to F4 using single seed descent (SSD) method. The selected single plants in F4 generation were raised as F5 progeny rows and harvested as F6 progeny bulks. During Rainy 2024, a total of 1645 F6 progeny bulks were screened simultaneously in LeasyScan and in artificial foliar fungal disease (FDR) screening nursery (<5 in a 0 to 9 scoring scale at 75 days after planting). A total of 322 F6 progeny bulks from 104 families were selected and advanced to MSE testing. ICGV 171044, one of the parents in the 12 selected families was identified as a potential genotype contributing to early vigour. The mid-density genotyping panel containing 2500 SNP markers distributed across 20 groundnut chromosomes from 263 cultivated accessions is being used to genotype all the lines advanced to Stage I multi-environment testing. The 1645 progeny bulks are genotyped using 5K mid-density array used in developing Genomic Selection models for early vigor and late leaf spot disease resistance. These progenies will be categorized into training set and testing set for GS model development and crossvalidation studies.