Skip to main content
. 2013 Sep 23;14:644. doi: 10.1186/1471-2164-14-644

Figure 5.

Figure 5

Principal component analysis of Sowetan cases and controls recruited for a rheumatoid arthritis association study. To assess the validity of our own sample as a suitable reference for the black Sowetan population, we used PCA to compare results with independently recruited case (SCA; n = 304) and control (SCO; n = 318) samples selected for a rheumatoid arthritis study. Cases and controls matched the clustering pattern of BSO individuals, supporting the use of the latter as a reference sample. However, a minority of cases displayed wider dispersal, with several individuals positioned more closely to Indian and European populations. Plot was based on a panel of 21 412 SNPs. BSO - Southeastern Bantu-speakers from the Soweto region; CEU - Utah residents of European ancestry; GIH - Gujarati Indians from Houston, Texas; LWK - Luhya in Webuye, Kenya; MKK - Maasai in Kinyawa, Kenya; SCA - Black Sowetan case individuals with rheumatoid arthritis; SCO - Black Sowetan control individuals; SEB - Southeastern Bantu-speakers; YRI - Yoruba in Ibadan, Nigeria.