Abstract for presentation at 11th International Congress of Human Genetics

Population genetic structure, ancestry, admixture and confounding in the Reynold-Stanford cardiovascular case-control association study

  • Analabha Basu, University of California, San Francisco, India
  • Thomas Quertermous, Stanford University, United States
  • Mark Hlatky, Stanford University, United States
  • Richard Myers, Stanford University, United States
  • Carlos Iribarren, Kaiser Permanente of Northern California, Oakland, United States
  • Alan Go, Kaiser Permanente of Northern California, Oakland, United States
  • Malini Chandra, Kaiser Permanente of Northern California, Oakland, United States
  • Neil Risch, University of California, San Francisco, United States

    • Donald W. Reynolds Cardiovascular Research Center at Stanford University, in collaboration with the Kaiser Permanente Division of Research, is focusing on the study of the genetic basis of cardiovascular disease in multiple ethnic populations. The study included the analysis of 467 genetic variants within 77 candidate genes. From these data we were able to address questions relating to the relationship between an individual's race/ethnicity self-identification and genetic clustering. Based on self reported grandparental ancestry information, the study also allowed us to specifically look at the question of admixed ancestry, especially for individuals who self identified into multiple ethnicities.
    Genetic cluster analysis produced five major clusters with near-perfect correspondence with the self-reported Caucasoid, African-American, Hispanic, South-Asian and East Asian populations. For all pairwise comparisons the South Asians and the Hispanics were genetically the closest groups having the minimum distance between them. They were virtually inseparable in two-dimensions, but the South-Asians clearly separate out as we added the third dimension in the Multi-dimensional scaling plot. As a group the South-Asians were distinct from both East Asians and Caucasoids, with whom they are often mistakenly clubbed in genetic analysis. Their genetic similarity to Hispanics sheds some light on deep rooted continental ancestry. We also found self-reported grandparental information as a powerful tool to classify individuals having mixed ancestry. Using the ‘admixture model’ option of the software ‘structure’, we demonstrate how self reported grandparental information corresponds closely to genetic estimates of an individual’s ancestry.

    Conference Organiser - ICMS Pty Ltd