Extended Data Fig. 10: Shannon Diversity Index scores for var transcript diversity for all single cell transcriptomes obtained using HIVE. | Nature Microbiology

Extended Data Fig. 10: Shannon Diversity Index scores for var transcript diversity for all single cell transcriptomes obtained using HIVE.

From: scRNA-seq reveals transcriptional plasticity of var gene expression in Plasmodium falciparum for host immune avoidance

Extended Data Fig. 10

(a) The Shannon Diversity Index for var gene expression was determined for each single cell transcriptome obtained by HIVE for each of the populations shown on the horizontal axis. Cells categorized as ‘single’ or ‘many’ are displayed in the columns marked S and M, respectively. The Shannon Index was calculated based on the number of UMIs for each var gene within the transcriptome of each cell using the formula \({H}^{{\prime} }=-{\sum }_{i=1}^{S}{p}_{i}{\log }_{2}{p}_{i}\). In this formula, H’: Index score; i: each category, in this case each var gene; S: number of categories, in this case the number of var genes; pi: proportion of the total made up of category i. Cells in which transcripts were detected from only one var gene have no diversity and a score of zero. A diversity score of 0.175 (shown as a red horizontal line) separates the vast majority of ‘single’ cells from ‘many’ cells. There is overlap however, particularly in samples with exceptional sequencing depth when very low levels of reads could be detected from nearly silent genes (for example the 3D7 High Single A population). In such instances, a minority of cells that we categorized as ‘singles’ display a Shannon Index score similar to ‘multiples’, however in these cells only one gene reaches the 15% threshold, a property that we believe is biologically relevant for the reasons described in detail in the Methods section and that is not reflected in the Shannon Index score. (b) Table showing the average index score for each category and the number of cells scoring either above or below the cutoff of 0.175.

Source data

Back to article page