Figure 4

Patterns of admixture and population structure in the Central European historical wheat collection. (a) STRUCTURE model with K = 2 optimal clustering. Each accession is represented by an individual vertical line divided into K colored segments with heights according to genotype memberships in the clusters. The K = 2 divided the collection to landraces (Q1) and modern varieties (Q2). (b) At K = 3, the cluster of landraces was separated into two groups (Q1, Q2), and modern varieties were placed alongside closely related landraces (Q3). (c) Accessions and K = 6 clustering, vertical lines were given membership coefficients for each of the six clusters (Q1–6) if they had 50% or more participation in that group. (d) Plotted are STRUCTURE models with indicated K = 6 optimal clustering Q groups sorted according to chronological origin (landraces vs modern cultivars). (e) Pie charts depicting the average individual membership proportions (Q) in each of the six inferred ancestral groups identified by STRUCTURE analysis for each region from which landraces were gathered. The landraces were dominated by four ancestral groups (Q1, 3, 4, and 6) supporting the distinct separation of landraces from modern elite varieties. (f) Pie charts depicting the percentage memberships of modern wheat varieties. Ancestral group 5 dominates modern cultivars, while group 2 and 5 are mixed in Central Europe collection. The chart shows how the previously dominant four ancestral groups (Q1, 3, 4, and 6) in landraces were replaced by group Q2 and 5 over time in modern varieties.