Ethnic Variation in Inflammatory Profile in Tuberculosis
Figure 2
Principal component analysis (PCA) plots generated using immunological parameters which contribute to variation in baseline inflammatory profile between PTB patients of African and Eurasian ancestry.
Each point represents one patient, and its position in the plot is determined by the combined effects of all parameters measured for that patient sample that contribute significantly to ethnic variation in inflammatory profile. The distance between sample points represents Euclidean distance. The first 3 component vectors are displayed, along with a % figure signifying the proportion of the variability in the data that each component accounts for. Points representing patients with Eurasian ancestry have been coloured according to their ethnic subgroup (Central/South Asian (blue) and European/Middle Eastern (yellow) ancestry) to demonstrate that they cluster together and that they are separated from those representing patients of African ancestry (black). One patient of mixed South Asian and European ancestry is classified as a Eurasian (green) and clusters within samples from the Eurasian subgroups. A, PCA plot of circulating immunological parameters in patients of African (n = 45), Central/South Asian (n = 55), European/Middle Eastern (n = 27) and Eurasian (n = 1) ancestry at baseline. B, PCA plot of rCFP-10-stimulated immunological parameters in patients of African (n = 13), Central/South Asian (n = 22) and European/Middle Eastern (n = 7) ancestry at baseline.