ReMap Database

Overview

ReMap 2022 is an integrative database of transcription factor (TF), cofactor, and chromatin regulator binding sites derived from uniformly reprocessed ChIP-seq experiments. The 2022 release catalogs 165 million non-redundant peaks from 8,113 ChIP-seq datasets covering 1,210 TFs across human (hg38/hg19), mouse (mm10), Drosophila, and Arabidopsis genomes. All peaks are called with a consistent pipeline from public GEO/ArrayExpress experiments. Access is via the ReMap 2022 REST API at https://remap2022.univ-amu.fr/api/ and bulk BED file downloads; no authentication required.

When to Use

• Finding all TFs with ChIP-seq peaks overlapping a genomic region of interest (e.g., a GWAS SNP locus or candidate enhancer)
• Retrieving TF peaks near a gene's transcription start site to map its proximal regulatory landscape
• Listing all TFs available in ReMap for human or mouse with their peak and dataset counts
• Filtering ChIP-seq peaks by regulatory biotype annotation (promoter, enhancer, exon, intron, intergenic) for a TF in a specific cell line
• Downloading a BED file of all binding peaks for a TF across all cell types for offline analysis
• Identifying co-binding TFs at a locus by querying all overlapping peaks and grouping by TF name
• Use jaspar-database instead when you need PWM/PFM sequence models of TF binding specificity rather than ChIP-seq peak locations
• For ENCODE-specific regulatory tracks and accessibility data use encode-database; ReMap aggregates TF binding peaks from many sources including ENCODE

Prerequisites

• Python packages: requests, pandas, matplotlib
• Data requirements: genomic coordinates (GRCh38/hg38 or hg19), gene names, or TF names
• Environment: internet connection; no API key required
• Rate limits: no official published limits; use time.sleep(0.5) between batch requests to avoid server overload
• Note: The ReMap API is a research API; endpoint availability may vary. All examples include a BED download fallback.

pip install requests pandas matplotlib

Quick Start

import requests

REMAP_API = "https://remap2022.univ-amu.fr/api/v1"

# Query TF peaks overlapping a genomic region
r = requests.get(f"{REMAP_API}/peaks/overlap/", params={
    "chr": "chr17",
    "start": 7_670_000,
    "end": 7_690_000,
    "assembly": "hg38"
}, timeout=30)
r.raise_for_status()
peaks = r.json()
print(f"Peaks overlapping TP53 locus: {len(peaks)}")
tfs = set(p.get("name", "").split(":")[0] for p in peaks)
print(f"Unique TFs: {len(tfs)}")
print(f"TF names (first 10): {sorted(tfs)[:10]}")

Core API

Query 1: Region Overlap

Find all TF ChIP-seq peaks overlapping a specified genomic window. Returns peak records including TF name, cell type, coordinates, and score.

import requests, time, pandas as pd

REMAP_API = "https://remap2022.univ-amu.fr/api/v1"

def query_region(chrom, start, end, assembly="hg38", timeout=30):
    """Return all ReMap peaks overlapping [chrom:start-end]."""
    r = requests.get(f"{REMAP_API}/peaks/overlap/", params={
        "chr": chrom, "start": start, "end": end, "assembly": assembly
    }, timeout=timeout)
    r.raise_for_status()
    return r.json()

# Query 100 kb window on chr17 around TP53
peaks = query_region("chr17", 7_670_000, 7_690_000, assembly="hg38")
print(f"Total peaks: {len(peaks)}")

# Parse name field: format is "TF:experiment_id:cell_type"
rows = []
for p in peaks:
    parts = p.get("name", "::").split(":")
    tf   = parts[0] if len(parts) > 0 else ""
    exp  = parts[1] if len(parts) > 1 else ""
    cell = parts[2] if len(parts) > 2 else ""
    rows.append({
        "chr": p.get("chr", p.get("chrom", "")),
        "start": p.get("start", 0),
        "end": p.get("end", 0),
        "tf_name": tf,
        "experiment_id": exp,
        "cell_type": cell,
        "score": p.get("score", 0),
    })

df = pd.DataFrame(rows)
print(f"\nUnique TFs: {df['tf_name'].nunique()}")
print(f"Top TFs by peak count:\n{df['tf_name'].value_counts().head(10).to_string()}")

# Fallback: if API is unavailable, use a locally downloaded BED file
# Download from: https://remap2022.univ-amu.fr/download_page
# e.g., remap2022_all_macs2_hg38_v1_0.bed.gz

import pandas as pd

def query_region_from_bed(bed_file, chrom, start, end):
    """Filter a ReMap BED file for overlapping peaks."""
    cols = ["chr", "start", "end", "name", "score", "strand",
            "thick_start", "thick_end", "color"]
    df = pd.read_csv(bed_file, sep="\t", header=None, names=cols,
                     compression="infer")
    mask = (df["chr"] == chrom) & (df["end"] > start) & (df["start"] < end)
    return df[mask].reset_index(drop=True)

# Usage (requires downloaded BED):
# df = query_region_from_bed("remap2022_all_macs2_hg38_v1_0.bed.gz",
#                             "chr17", 7_670_000, 7_690_000)

Query 2: Gene-Centric Query

Retrieve all TF ChIP-seq peaks near a gene's TSS, providing a promoter-proximal regulatory landscape for the gene.

import requests, time, pandas as pd

REMAP_API = "https://remap2022.univ-amu.fr/api/v1"

def query_gene_peaks(gene_name, assembly="hg38", timeout=30):
    """Return all ReMap peaks near a gene TSS."""
    r = requests.get(f"{REMAP_API}/peaks/gene/", params={
        "gene": gene_name, "assembly": assembly
    }, timeout=timeout)
    r.raise_for_status()
    return r.json()

peaks = query_gene_peaks("MYC", assembly="hg38")
print(f"Peaks near MYC TSS: {len(peaks)}")

rows = []
for p in peaks:
    parts = p.get("name", "::").split(":")
    rows.append({
        "tf_name": parts[0] if parts else "",
        "cell_type": parts[2] if len(parts) > 2 else "",
        "chr": p.get("chr", p.get("chrom", "")),
        "start": p.get("start", 0),
        "end": p.get("end", 0),
        "score": p.get("score", 0),
        "biotype": p.get("biotype", ""),
    })

df = pd.DataFrame(rows)
print(f"\nTFs near MYC TSS ({df['tf_name'].nunique()} unique):")
print(df["tf_name"].value_counts().head(10).to_string())
print(f"\nCell types represented: {df['cell_type'].nunique()}")

Query 3: TF Browser

List all TFs available in ReMap for a given genome assembly, with peak and experiment counts.

import requests, time, pandas as pd

REMAP_API = "https://remap2022.univ-amu.fr/api/v1"

def list_tfs(assembly="hg38", timeout=30):
    """Return all TFs in ReMap for the given assembly with statistics."""
    r = requests.get(f"{REMAP_API}/tfbs/list/", params={"assembly": assembly}, timeout=timeout)
    r.raise_for_status()
    return r.json()

def get_database_stats(assembly="hg38", timeout=30):
    """Return overall database statistics for the assembly."""
    r = requests.get(f"{REMAP_API}/stats/", params={"assembly": assembly}, timeout=timeout)
    r.raise_for_status()
    return r.json()

# Database overview
try:
    stats = get_database_stats("hg38")
    print(f"ReMap 2022 hg38 statistics:")
    for k, v in stats.items():
        print(f"  {k}: {v}")
except Exception as e:
    print(f"Stats endpoint unavailable: {e}")
    print("ReMap 2022 hg38: 165M peaks, 1,210 TFs, 8,113 datasets (from publication)")

# TF list
try:
    tfs = list_tfs("hg38")
    df_tfs = pd.DataFrame(tfs)
    print(f"\nTFs available (hg38): {len(df_tfs)}")
    if "peak_count" in df_tfs.columns:
        top = df_tfs.nlargest(10, "peak_count")[["name", "peak_count", "dataset_count"]]
        print("Top 10 TFs by peak count:")
        print(top.to_string(index=False))
except Exception as e:
    print(f"TF list endpoint unavailable: {e}")
    print("Use TF name queries directly (Query 4) or download TF-specific BED files.")

Query 4: TF-Specific Peak Query

Retrieve all peaks for a named TF in a given assembly, optionally filtered by cell type.

import requests, time, pandas as pd

REMAP_API = "https://remap2022.univ-amu.fr/api/v1"

def query_tf_peaks(tf_name, assembly="hg38", timeout=30):
    ""