Peruvian maize exhibits significant morphological diversity, with
landraces cultivated from sea level up to 3,500 meters above sea level.
Previous research based on morphological descriptors identified at least
52 Peruvian maize races, but their genetic diversity and population
structure remain largely unknown. In this study, we used
genotyping-by-sequencing (GBS) to infer the genetic structure and
diversity of 423 maize accessions from the Genebank of La Molina National
Agrarian University (UNALM). These accessions represent nine races and one
sub-race, along with 15 open-pollinated lines (purple corn) and two yellow
maize hybrids. We obtained 14,235 high-quality SNPs distributed along the
10 maize chromosomes. Gene diversity ranged from 0.33 (Pachia) to 0.362
(Ancashino), with Cusco showing the lowest inbreeding coefficient (0.205)
and Ancashino the highest (0.274) among the landraces. Population
divergence (FST) was very low (mean = 0.017), indicating extensive
interbreeding among Peruvian maize varieties. Population structure
analysis revealed that these 423 distinct genotypes could be grouped into
10 clusters, with some maize races clustering together. Peruvian maize
races did not form monophyletic groups; instead, our phylogenetic tree
identified two clades corresponding to the chronological classification of
Peruvian maize races: Anciently Derived or Primary Races (ADPR) and Lately
Derived or Secondary Races (LDSR). These clades also align with the
geographic origins of the maize races, reflecting their mixed evolutionary
backgrounds. Further investigation of Peruvian maize germplasm using
modern technologies is essential to enhance their use in breeding
programs, particularly in the Andean region of Peru.