Comparison of genetic diversity using morphometric, molecular and digital imaging parameters in castor (Ricinus communis L.)

Abstract

Castor (Ricinus communis L.) is highly remunerative industrially important non-edible oilseed crop with immense and diversified usage. A study was conducted with 26 diverse castor genotypes (20 pistillate and six monoecious) to evaluate diversity using 17 morphometric characters and 22 SSR (simple sequence repeat) markers. Cluster analysis using both hierarchical (Euclidean distance matrix using the Ward method) and non-hierarchical (K means) method grouped the castor lines into five clusters. The genotypes YTP 1, JP 65, DPC 19 and TMV 5 were genetically wide and could be explored further during cross breeding to generate superior recombinants. Principal component analysis based on eigen vectors revealed that the first six components cumulatively explained 86.17% of the total variation. Association analysis exhibited positive relation among seed yield attributing traits. Using 22 microsatellite markers 16 (72.72 %) were polymorphic. The molecular weight (bp) of the amplified products differed from 120 (RCM 10) to 300 bp (RCM 7). A total number of alleles/loci were 63 alleles with the average of 3.94. With a mean value of 0.57, the highest gene diversity (0.75) was for RCM 12 while it was lowest (0.28) for RCM 43. Based on PIC (polymorphic information content) values, 0.26 (RCM 43) to 0.71 (RCM 12), 10 markers were highly informative. The pistillate lines Geeta, M 574 and DPC 24 were diverse and placed in distinct clusters. The findings revealed that both phenotypic and molecular markers should be deployed to capture the actual genetic diversity and to identify promising genotypes to reap maximum heterosis through hybridization.