HaploReg v5 documentation

Back to HaploReg v5 | Full documentation (redesigned site)

Contents

Overview

HaploReg helps turn a list of disease- or trait-associated variants into mechanistic hypotheses. Because a GWAS signal usually tags a whole haplotype rather than the causal variant itself, HaploReg first expands each query variant to the variants in linkage disequilibrium with it, then annotates every candidate with evidence that it might be functional: chromatin state across reference epigenomes, DNA accessibility, bound proteins, altered regulatory motifs, evolutionary constraint, known trait and expression associations, and its position relative to genes.

A typical reading of a results table: look for variants that sit in an enhancer or promoter (orange / red cells) in a tissue relevant to the trait, that alter a motif for a factor active in that tissue (pink), and that are bound by proteins (green) or open chromatin (grey). eQTL hits (yellow) suggest which gene is affected.

Quick start

1. On the Build Query tab, enter rsIDs (e.g. rs7903146), or a region such as chr10:112990000-113000000, or upload a text file with one rsID per line, or choose a curated GWAS locus set.
2. On Set Options, pick the LD threshold (r²) and the population used to compute LD, and any display options.
3. Click Submit. Each query SNP gets a table of variants in LD. Query SNPs are shown in red.
4. Hover over a colored cell for the full list behind it. Click a variant to open its detail view in a new window.
5. Choose Text output to get tab-separated values for analysis.

Results table columns

Each column links to its full definition below.

Reference tables

ChromHMM core 15-state model

ChromHMM 25-state model (imputed marks)

Tissue codes

Text output columns

Reference epigenomes (E001–E129)

The HaploReg v5 pipeline

Version 5 rebuilds the HaploReg database from the 1000 Genomes high-coverage (30×, GRCh38) call set. For each chromosome the pipeline keeps #CHROM, POS, ID, REF, ALT and the per-superpopulation allele frequencies in unrelated samples (AF_*_unrel), assigns rsIDs, annotates every variant with its ChromHMM chromatin state in each reference epigenome, and scans reference and alternate alleles against the motif library with MOODS. Per-chromosome VCFs are bgzipped, tabix-indexed and merged. Chromosome Y has no ChromHMM annotation, so its chromatin-state fields are written as . to keep the header identical across chromosomes.

The web interfaces read data through a small adapter layer (see README.md), so the pipeline output can replace the interim Ensembl source without changing either interface.

Differences from HaploReg v4

• Annotations are computed natively on GRCh38 (v4 annotated on hg19 and displayed hg38 positions).
• Five superpopulations (AFR, AMR, EAS, EUR, SAS) replace v4's four. v4's ASN contained only East Asian samples and corresponds to EAS; South Asian (SAS) populations are new.
• The default conservation column is GERP, which the interim live source can supply.
• A Data source option chooses between live Ensembl data and the offline synthetic demo.
• Every column and option links to this documentation.

References

Contact: haploreg@mit.edu