Fig. 1: Diversity panel of wild and domesticated barley.

a, Collection sites and population structure of 143 wild barley genotypes with precise geographical locations. The coloured dots show the results of model-based ancestry estimation with ADMIXTURE (the number of ancestral populations (K) = 5, predominant ancestry component) and are plotted at approximate collection sites. Jitter was added to avoid overlaps between nearby accessions. Only unadmixed samples, that is, those whose major ancestry components was 0.85 or more are shown. CA, Central Asia; NL, northern Levant; NM, northern Mesopotamia; SD, Syrian desert; SL, southern Levant. b, Assignment to macrogeographical regions of 15 populations inferred from GBS data of 19,778 domesticated barley genotypes²⁴. The population names encode the most common origin of samples and their predominant morphological and phenological characters (row type, lemma adherence and annual growth habit) as detailed in Supplementary Table 6. c, Archaeological sites at which ancient barley grains used for ancient DNA extractions were found. Ages of the samples, as determined by radiocarbon dating, are indicated in the figure. Geographical outlines in panels a–c were obtained from the R package ‘maps’ (https://CRAN.R-project.org/package=maps), which uses public domain base map data (under a GNU General public license: version 2).