Table 1 . HapMap 3 panels used for cross-validation.
| Panel ID | Panel Description | Number of Unrelated Individualsa |
|---|---|---|
| ASW | African ancestry in Southwest USA | 63 |
| CEU | Utah residents with Northern and Western European ancestry from the CEPH collection | 117 |
| CHBb | Han Chinese in Beijing, China | 84 |
| CHD | Chinese in Metropolitan Denver, Colorado | 85 |
| GIH | Gujarati Indians in Houston, Texas | 88 |
| JPTb | Japanese in Tokyo, Japan | 86 |
| LWK | Luhya in Webuye, Kenya | 90 |
| MKK | Maasai in Kinyawa, Kenya | 143 |
| MXL | Mexican ancestry in Los Angeles, California | 52 |
| TSI | Toscani in Italia | 88 |
| YRI | Yoruba in Ibadan, Nigeria | 115 |
| All panels | 1011 |
a
In panels that included trios (ASW, CEU, MXL, MKK, and YRI), we retained the trio parents as “unrelated” individuals. In panels that included parent-child duos (ASW, CEU, MXL, and YRI), we retained the observed duo parent and the inferred transmitted haplotype from the unobserved duo parent, yielding three “unrelated” haplotypes per duo; we then paired the inferred transmitted haplotypes at random to create diploid pseudo-individuals.
b
We combined the CHB and JPT panels into a single CHB+JPT panel with 170 individuals for all of the analyses in this paper.