Fig. 2: Validation of ChromA on ground-truth data sets. | Nature Communications

Fig. 2: Validation of ChromA on ground-truth data sets.

From: Characterizing chromatin landscape from aggregate and single-cell genomic assays using flexible duration modeling

Fig. 2

a Genomic loci were selected to create a validation data set from highly expressed and transcription factor genes regulating Th17 development. To manually curate genomic regions, information from ChIP-seq and ATAC-seq experiments were combined. b Example of curated genomic locus flanking the Irf4 gene in the mouse genome. ATAC-seq, ChIP-seq, PeaKDEck (PD), MACS2 (M2), ChromA, and manual annotations are displayed. ChromA algorithm recovers a greater number of ground-truth peaks than competing algorithms and covers each peak more thoroughly. c Fraction of manually annotated peaks covered with at least one peak. d Average fraction of peaks covered per genomic loci (mean +/− s.e.m., the number of regions equals 10). e F1 score.

Back to article page