DEVELOPMENT OF THE CROPPOPSTRUCTURE GENOMIC PLATFORM FOR GENETIC DIVERSITY ANALYSIS AND SUPPORT OF BREEDING DECISIONS

Understanding genetic diversity and population structure is essential for efficiently using crop germplasm collections in breeding, yet a complete analysis requires many separate programs and command-line skills. This study aimed to develop CropPopStructure, a web-based genomic platform that performs a full population genetic analysis of a germplasm collection from a VCF file and a population list, and to validate it on data with a known structure. The platform combines a Nextflow pipeline based on PLINK, VCFtools, and R packages with an R Shiny web interface accessible from a browser. It validates the input data and runs eleven analysis modules: genetic diversity indices, principal component analysis (PCA), discriminant analysis of principal components (DAPC), sNMF ancestry estimation, genomic kinship, a neighbor-joining tree, analysis of molecular variance (AMOVA), pairwise FST, private SNPs, SNP density, and linkage disequilibrium (LD) decay. HapMap files can be converted to VCF with TASSEL within the platform. The platform was tested on a simulated collection of 98 homozygous lines from four populations genotyped with 35,000 SNPs and generated under a hierarchical coalescent model. PCA separated all four populations (PC1 = 6.32%, PC2 = 1.96% of variance), and the BIC criterion indicated two main genetic groups corresponding to the model's two lineages. AMOVA attributed 21.9 % of molecular variance to differences among populations (ΦPT = 0.219; p < 0.001). Pairwise FST values were 0.037–0.039 between sister populations and 0.124–0.125 between lineages, reproducing the simulated hierarchy. CropPopStructure allows breeders and geneticists without bioinformatics training to obtain reproducible results that support forming core collections, selecting genetically distant parents, and controlling population structure in association studies.