Fig. 2
The H3K27ac landscape undergoes global changes in stunted children within their first year of life. a Overview of the experimental design. Each line in the plot with HAZ scores represents the change of HAZ score from birth to 1 year of age for a child whose PBMC sample was used for the study. The left panel shows HAZ scores for 18-week-old children (n = 14), the right panel for 1-year-old children (n = 22). Red dots indicate HAZ < − 2 at a given age. Lines are colored based on ΔHAZ between birth and 18 weeks or between birth and 1 year. Bottom panel illustrates three different scenarios for differential analysis in which blue regions are downregulated in stunting, grey regions unchanged or with no significant trend; and red regions are upregulated in stunted children. b MA-plot shows changes in H3K27ac regions in 18-week-old children as the ΔHAZ (18 wk) score increases. Each dot is an H3K27ac region, the x-axis represents the mean read coverage over the region, and the y-axis indicates the log2(fold-change) of read coverage per unit increase of ΔHAZ (18 wk) score. Colored dots indicate significantly affected regions with false discovery rate (FDR) corrected p value < 0.05. c Same as b for 1-year-old children. d Alluvial plot showing changes to log2(fold-change) of corresponding regions between 18 weeks and 1 year. “Up”/“down”—H3K27ac levels are increased/decreased, accordingly, with higher risk of stunting at a given age, “not-present”—region was not identified at the given age