Figure 5.

DNA Motifs Enriched in Stage-Specific ATAC Regions

(A) Co-clustering of ATAC-seq and RNA-seq data, limited to peak-to-gene pairs with Pearson correlation > 0.6 (n = 2,118; Table S1). Accessibility and transcript abundance are expressed as proportion of sum of (qn)RPKM values over the time points and clustered by k-means using the 1-Pearson correlation distance metric into eight clusters. Color scales range from the 20^th to 80^th percentile per dataset.

(B) Heatmap of significance estimates for differential motif enrichment (expressed as −log10(p value)). Each column relates to a cluster generated by co-clustering the ATAC-seq and RNA-seq data by k-means cluster (see A). Each row refers to a motif or a “motif group,” with logo and name listed on the right. Asterisks indicate that the motif is a representative from a group of similar motifs (Figure S3B; Table S3). Predicted binding sites for P. falciparum AP2 TFs are reported in blue font with PlasmoDB geneID (name, if known, is in brackets). When a cluster contains one or more predicted P. falciparum AP2 motifs, these are reported in brackets behind the representative motif.

See also Figure S3 and Tables S1 and S3.