multi-omics-integration

What it does

Layer RNA-seq, ATAC-seq, Histone ChIP-seq, and TF ChIP-seq data from ENCODE to build a comprehensive regulatory landscape for a tissue or cell type, following Mawla et al. 2023 framework for cross-assay integration. Handles chromatin state annotation (ChromHMM), enhancer-gene linkage, TF motif enrichment, and cell type-specific regulatory element identification. Designed for ANY multi-omic analysis, enhancer discovery, regulatory network construction, or epigenomic characterization using ENCODE data.

How it works

The skill guides the user to define regulatory questions, then inventories available data across data layers (RNA-seq, ATAC-seq, Histone ChIP-seq for marks such as H3K27ac, H3K4me1, H3K4me3, H3K27me3, H3K36me3, and TF ChIP-seq). It prescribes search commands like encode_search_experiments and encode_list_files to gather data, produces a data availability matrix, and enforces quality gates per assay (e.g., mapping rate, FRiP, NSC/RSC/NRF). It then outlines integration approaches: ChromHMM for chromatin states, enhancer identification via convergence of signals (H3K27ac+, H3K4me1+, ATAC+, and NOT H3K4me3+ with promoter exclusion), defining poised enhancers (H3K4me1+ and H3K27ac-), and using ROSE for super-enhancers. It also details enhancer-gene linkage methods (nearest-gene, ABC Model, correlation, Hi-C, ENCODE links) and caveats about nearest-gene limitations. The stepwise workflows reference explicit data layers and state specific combinations for enhancer discovery and regulatory network inference.

When to use it

Use when you want to integrate multiple ENCODE data types to build a regulatory landscape for a tissue or cell type, especially to answer questions about active enhancers, promoters, and TF binding, and how these converge to define cell identity. The workflow is triggered by user questions about multi-omics integration, regulatory landscapes, enhancer discovery, or regulatory network construction.

What it can touch

It prescribes interacting with ENCODE data via commands:

• encode_search_experiments(…) across RNA-seq, ATAC-seq, Histone ChIP-seq, and TF ChIP-seq data
• encode_list_files(…) to download RNA-seq gene quantifications, ATAC-seq/Histone ChIP-seq IDR-thresholded peaks, and TF ChIP-seq peaks
• References data types like narrowPeak, broadPeak, and BigWig files; applies ENCODE Blacklist filtering

Caveats

Addresses data quality gates per assay (mapping rate, FRiP, NSC/RSC/NRF, IDR). Highlights that multi-omics integration strength depends on data availability across layers; missing ATAC-seq or histone marks reduces reliability of enhancer validation. No outcomes promised beyond the described integration logic and data handling steps.