Genome-wide genotyping data renew knowledge on genetic diversity of a worldwide alfalfa collection and give insights on genetic control of phenology traits

• DOI: 10.3389/fpls.2023.1196134
• Publication Date: 2023-07-05T17:24:01Z
• Authors: Marie Pégard; Philippe Barre; Sabrina Delaunay; Fabien Surault; Djura Karagić; Dragan Milić; Miroslav Zorić; Tom Ruttink; Bernadette Julier
• Keywords: 2. Zero hunger; PENALIZED SPLINES; LINKAGE DISEQUILIBRIUM; [SDV]Life Sciences [q-bio]; R-PACKAGE; Biology and Life Sciences; Plant culture; MOLECULAR MARKERS; genetic diversity; Plant Science; phenology; SB1-1110; REGULARIZATION PATHS; MIXED-MODEL; AUTOTETRAPLOID ALFALFA; FALL DORMANCY; GWAS; POPULATION-STRUCTURE; WINTER HARDINESS; alfalfa genetic diversity phenology GWAS genomic prediction; alfalfa; genomic prediction

China's and Europe's dependence on imported protein is a threat to the food self-sufficiency of these regions. It could be solved by growing more legumes, including alfalfa that highest producer under temperate climate. To create productive high-value varieties, use large genetic diversity combined with genomic evaluation improve current breeding programs. study diversity, we have used set 395 accessions (i.e. populations), mainly from Europe, North South America China, fall dormancy ranging 3 7 scale 11. Five breeders provided materials (617 accessions) were compared 400 accessions. All genotyped using Genotyping-by-Sequencing (GBS) obtain SNP allele frequency. These data describe identify groups. The phenotyped for phenology traits (fall flowering date) at two locations (Lusignan in France, Novi Sad Serbia) 2018 2021. QTL detected Multi-Locus Mixed Model (mlmm). Subsequently, quality prediction each trait was assessed. Cross-validation assess testing GBLUP, Bayesian Ridge Regression (BRR), Lasso methods. A structure seven groups found. Most related geographical origin showed European American material genetically distinct Chinese material. Several associated found most linked genes. In our study, infinitesimal methods higher than Lasso, achieved high (>0.75) predicting abilities some cases. Our results are encouraging showing it possible achieve quality.