Genomic insights into <em>Lolium multiflorum</em> diversity for forage breeding in Andean livestock systems

Understanding the genetic diversity of Lolium multiflorum is essential for strengthening breeding programs and enhancing the sustainability of Andean livestock systems. This study evaluated the genomic variability and population structure of 27 accessions of Lolium multiflorum from the Cajamarca region and the INIA Amazonas germplasm bank (Peru), using the genotyping-by-sequencing (GBS) technique. DNA extracted from young leaves was sequenced on an Illumina NovaSeq 6000 platform, and reads were processed with bioinformatic tools (FastQC, BWA, GATK, VCFtools, BCFtools, and scikit-allel). After stringent filtering, 2,070 SNPs were obtained across seven chromosomes, with heterogeneous SNP distribution across the seven chromosomes. Principal Coordinate Analysis (PCoA) and phylogenetic analysis (UPGMA) revealed two distinct genetic groups, indicating a complex structure shaped by gene flow and local selection. AMOVA showed that 90.01% of the genetic variation occurs within populations, whereas 9.99% corresponds to interregional differences (PhiPT = 0.099, p < 0.006). The negative FIS values (Cajamarca = -0.2312; Amazonas = -0.5489) indicate an excess of heterozygotes, a pattern typically associated with predominantly allogamous species. Moreover, the high levels of observed heterozygosity (Ho > 0.57) point to possible hybrid vigor and suggest that these populations may be maintaining a stable genetic equilibrium. These results confirm that Peruvian L. multiflorum maintains a broad genetic base, shaped by historical germplasm exchange and local environmental adaptation. This genetic diversity provides essential insights for conservation planning and supports breeding initiatives to improve forage resilience and productivity in high-Andean ecosystems. Methods Leaf tissue samples from Lolium multiflorum accessions collected in the northern Peruvian Andes were used for genomic analysis. Genomic DNA was extracted from young leaves using the NucleoSpin® Plant II kit following the manufacturer’s protocol. DNA quality and concentration were assessed by agarose gel electrophoresis and fluorometric quantification. Genotyping-by-sequencing (GBS) libraries were prepared using the ApeKI restriction enzyme and sequenced on an Illumina NovaSeq 6000 platform to generate paired-end reads. Raw sequencing reads were quality-checked using FastQC and filtered with fastp. Reads were aligned to the Lolium multiflorum reference genome (cultivar Rabiosa) using BWA, and SNP calling was performed with GATK following best-practice pipelines. Only high-quality biallelic SNPs were retained after filtering for read depth, quality metrics, minor allele frequency, missing data, and linkage disequilibrium. Biological replicates were merged to obtain a final non-redundant dataset of unique accessions. Population structure, genetic diversity, and phylogenetic relationships were analyzed using ADMIXTURE, PLINK, VCFtools, scikit-allel, and R-based statistical packages.