Data from: Population genetics and origin of the native North American agricultural weed waterhemp (Amaranthus tuberculatus)

Premise of the study: The evolution of invasiveness has been extensively studied in natural ecosystems; however, far less is known about the evolution of agricultural invasiveness, despite the major economic impact of weeds on crop productivity. Examining the population structure of recently arisen weeds can provide insights into evolutionary avenues to invasion of agroecosystems. Weeds that originate from wild plants are the most common yet least frequently studied type of agricultural invasive. Here we address several questions about the origin of the native North American agricultural weed waterhemp (Amaranthus tuberculatus), which invaded corn and soy fields in the midwestern United States in the 20th century. Methods: We genotyped 38 populations from across the species range with 10 microsatellite markers and used these data to assess genetic diversity and population structure within and outside the geographical region where waterhemp is agriculturally problematic. Key results: We found evidence for two ancestral genetic lineages in our data, supporting the hypothesis that A. tuberculatus was diverging into two evolutionary lineages prior to the 20th century. However, we found no support for the hypothesis that agricultural weed populations arose from admixture of these two lineages after secondary contact. Our data suggest that eastward movement of the western genetic lineage, facilitated by changing agricultural practices, is the source of the agricultural invasion of waterhemp. Conclusions: This research demonstrates that agricultural invasion by native, wild plant species can proceed via different evolutionary trajectories from weeds related to domesticated plants, which has implications for evolutionary biology and weed control.Microsatellite Data for Waselkov and Olsen 2014Microsatellite data for all 38 populations of Amaranthus tuberculatus and all 10 microsatellite loci used in Waselkov and Olsen 2014. Data is in STRUCTURE format, two lines per individual, with "-9" indicating missing data.SSR Data for Waselkov and Olsen 2014.tx