ChIP-Enrich: gene set enrichment testing for ChIP-seq data.
Bottom Line: Adjustment for gene locus length is necessary because it is often positively associated with the presence of one or more peaks and because many biologically defined gene sets have an excess of genes with longer or shorter gene locus lengths.We identify DNA-binding proteins, including CTCF, JunD and glucocorticoid receptor α (GRα), that show different enrichment patterns for peaks closer to versus further from transcription start sites.We also identify known and potential new biological functions of GRα.
Affiliation: Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI 48109, USA Biostatistics Department, University of Michigan, Ann Arbor, MI 48109, USA.Show MeSH
Related in: MedlinePlus
Mentions: We asked whether ChIP-Enrich could identify known and potential new biology of a well-characterized transcription factor, the GRα (47). Previous analysis identified 4392 peaks in A549 cells treated with 100-nM DEX (dexamethasone stimulates GR activity); only 4.7% of the peaks were within 1 kb of a TSS (Figure 4a). GO term enrichment testing yielded largely distinct subsets of significant (FDR ≤ 0.05) terms for ‘nearest TSS’ (195 terms) and ‘≤1 kb from TSS’ (72 terms) with only 16 overlapping terms (Figure 4b and d; Supplementary Table S5). The most significant terms (after collapsing similar terms) are shown in Table 3. Terms significant using one or both locus definitions include ‘epithelial cell differentiation’ (q-values: nearest TSS = 1.8 × 10−6; ≤1 kb from TSS = 1.0) and ‘negative regulation of blood coagulation’ (q-values: nearest TSS = 0.077; ≤1 kb from TSS = 3.19 × 10−7, with the related term ‘regulation of wound healing’ (q-values: nearest TSS = 0.0064; ≤1 kb from TSS = 0.0029). In addition, we observed ‘response to glucocorticoid stimulus’ (q-values: nearest TSS = 0.0035; ≤1 kb from TSS = 0.55) and ‘regulation of lipid metabolic process’ (q-values: nearest TSS = 0.0062; ≤1 kb from TSS = 0.74). GRα is known to be involved in the response to steroids and the activation of lipolysis (48,49), although knowledge of the transcriptional role of GRα in wound healing and blood coagulation is more limited. We also tested for enrichment using non-overlapping locus definitions for regions closer to a TSS (≤5 kb from TSS; 14.5% of peaks) and further from a TSS (>10 kb from TSS; 75.6% of peaks) and again identified largely distinct gene sets (Supplementary Figure S10).
Affiliation: Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI 48109, USA Biostatistics Department, University of Michigan, Ann Arbor, MI 48109, USA.