Regulomedb Database

pip install requests pandas matplotlib

import requests

BASE = "https://regulomedb.org"

def regulome_score(variant, genome="GRCh38"):
    """Score a single variant (rsID or chr:pos-pos) via the GET /regulome-search/ endpoint."""
    r = requests.get(
        f"{BASE}/regulome-search/",
        params={"regions": variant, "genome": genome, "format": "json"},
        timeout=30,
    )
    r.raise_for_status()
    d = r.json()
    rs = d.get("regulome_score", {})
    vs = d.get("variants", [])
    return {
        "query": variant,
        "ranking": rs.get("ranking"),           # 1a / 1b / ... / 7
        "probability": float(rs.get("probability", 0)),
        "rsids": vs[0].get("rsids") if vs else [],
        "chrom": vs[0].get("chrom") if vs else None,
        "pos": vs[0].get("start") if vs else None,
    }

print(regulome_score("rs4946036"))
# {'query': 'rs4946036', 'ranking': '7', 'probability': 0.18412,
#  'rsids': ['rs4946036'], 'chrom': 'chr6', 'pos': 114819799}

import requests

BASE = "https://regulomedb.org"

def score_variant(variant, genome="GRCh38"):
    """Return the regulome_score block and resolved coordinates."""
    r = requests.get(
        f"{BASE}/regulome-search/",
        params={"regions": variant, "genome": genome, "format": "json"},
        timeout=30,
    )
    r.raise_for_status()
    d = r.json()
    rs = d.get("regulome_score", {})
    vs = d.get("variants", [])
    feats = d.get("features", {})
    print(f"Variant   : {variant}")
    print(f"Resolved  : {vs[0]['chrom']}:{vs[0]['start']} ({', '.join(vs[0].get('rsids', []))})")
    print(f"Ranking   : {rs.get('ranking')}  prob={rs.get('probability')}")
    print(f"Features  : ChIP={feats['ChIP']} Chromatin_accessibility={feats['Chromatin_accessibility']} "
          f"QTL={feats['QTL']} Footprint={feats['Footprint']} PWM_matched={feats['PWM_matched']}")
    return d

# Strong-regulatory locus example
score_variant("chr11:5226739-5226740")
# Ranking: 1a (HBB beta-globin promoter, multi-evidence)

# Score by chromosomal position alone
score_variant("chr17:7670000-7670001")  # TP53 region

import requests, pandas as pd

BASE = "https://regulomedb.org"

def scan_region(chrom, start, end, genome="GRCh38", limit=200):
    """List variants in a region with their resolved positions and overlapping rsIDs."""
    r = requests.get(
        f"{BASE}/regulome-search/",
        params={"regions": f"{chrom}:{start}-{end}", "genome": genome,
                "format": "json", "limit": limit},
        timeout=60,
    )
    r.raise_for_status()
    d = r.json()
    variants = d.get("variants", [])
    print(f"Variants in {chrom}:{start}-{end}: {len(variants)} (total indexed = {d.get('total')})")
    rows = [{"rsids": ", ".join(v.get("rsids", [])),
             "chrom": v.get("chrom"),
             "start": v.get("start"),
             "end": v.get("end")} for v in variants]
    return pd.DataFrame(rows)

df = scan_region("chr11", 5226000, 5227000)
print(df.head(10).to_string(index=False))

import requests, pandas as pd

BASE = "https://regulomedb.org"

def evidence_rows(variant, genome="GRCh38"):
    r = requests.get(
        f"{BASE}/regulome-search/",
        params={"regions": variant, "genome": genome, "format": "json"},
        timeout=60,
    )
    r.raise_for_status()
    g = r.json().get("@graph", [])
    rows = []
    for row in g:
        bs = row.get("biosample_ontology") or {}
        rows.append({
            "method": row.get("method"),
            "target_label": row.get("target_label"),
            "biosample": bs.get("term_name"),
            "organ_slims": ", ".join(bs.get("organ_slims") or []),
            "dataset": row.get("dataset", "").split("/")[-2] if row.get("dataset") else None,
            "value": row.get("value"),
        })
    return pd.DataFrame(rows)

df_evidence = evidence_rows("chr11:5226739-5226740")
# Each method is one of: ChIP-seq, Histone ChIP-seq, ATAC-seq, DNase-seq,
# footprints, PWMs, chromatin state, eQTLs
print(df_evidence["method"].value_counts())

import requests, pandas as pd

BASE = "https://regulomedb.org"

def tf_binding(variant, genome="GRCh38"):
    r = requests.get(f"{BASE}/regulome-search/",
                     params={"regions": variant, "genome": genome, "format": "json"},
                     timeout=60)
    r.raise_for_status()
    rows = []
    for g in r.json().get("@graph", []):
        if g.get("method") != "ChIP-seq":
            continue
        bs = g.get("biosample_ontology") or {}
        rows.append({
            "tf": g.get("target_label"),
            "biosample": bs.get("term_name"),
            "classification": bs.get("classification"),
        })
    return pd.DataFrame(rows)

df_tfs = tf_binding("chr11:5226739-5226740")
print(f"TF ChIP-seq peaks overlapping query: {len(df_tfs)}")
print(df_tfs.groupby("tf").size().sort_values(ascending=False).head(10))

import requests, pandas as pd

BASE = "https://regulomedb.org"

def tissue_scores(variant, genome="GRCh38", top_n=10):
    r = requests.get(f"{BASE}/regulome-search/",
                     params={"regions": variant, "genome": genome, "format": "json"},
                     timeout=30)
    r.raise_for_status()
    ts = r.json().get("regulome_score", {}).get("tissue_specific_scores", {})
    s = pd.Series({k: float(v) for k, v in ts.items()})
    return s.sort_values(ascending=False).head(top_n)

print("Top tissues by regulatory probability for chr11:5226739-5226740:")
print(tissue_scores("chr11:5226739-5226740"))

import requests, pandas as pd

BASE = "https://regulomedb.org"

def nearby(variant, genome="GRCh38"):
    r = requests.get(f"{BASE}/regulome-search/",
                     params={"regions": variant, "genome": genome, "format": "json"},
                     timeout=30)
    r.raise_for_status()
    rows = []
    for s in r.json().get("nearby_snps", []):
        rows.append({
            "rsid": s.get("rsid"),
            "chrom": s.get("chrom"),
            "pos": s.get("coordinates", {}).get("gte"),
            "type": s.get("variation_type"),
            "maf": s.get("maf"),
        })
    return pd.DataFrame(rows)

df_nearby = nearby("rs4946036")
print(f"Nearby SNPs to rs4946036: {len(df_nearby)}")
print(df_nearby.head(10).to_string(index=False))

# rsID — resolved server-side to current-build coordinates
"rs4946036"

# Single-position range
"chr11:5226739-5226740"

# Wider region (returns multiple variants[] entries + larger @graph)
"chr11:5226000-5227000"

import requests, time, pandas as pd
import matplotlib.pyplot as plt

BASE = "https://regulomedb.org"

gwas_snps = ["rs7903146", "rs10811661", "rs1801282", "rs4946036",
             "rs2268177", "rs10830963", "rs1111875"]

records = []
for snp in gwas_snps:
    r = requests.get(f"{BASE}/regulome-search/",
                     params={"regions": snp, "genome": "GRCh38", "format": "json"},
                     timeout=30)
    r.raise_for_status()
    d = r.json()
    rs = d.get("regulome_score", {})
    feats = d.get("features", {})
    g = d.get("@graph", [])
    tfs = sorted({row["target_label"] for row in g
                  if row.get("method") == "ChIP-seq" and row.get("target_label")})
    records.append({
        "snp": snp,
        "ranking": rs.get("ranking"),
        "probability": float(rs.get("probability", 0)),
        "has_qtl": feats.get("QTL", False),
        "tf_count": len(tfs),
        "num_evidence_rows": len(g),
    })
    time.sleep(0.3)

df = pd.DataFrame(records).sort_values("probability", ascending=False)
print(df.to_string(index=False))
df.to_csv("gwas_regulatory_priority.csv", index=False)

fig, ax = plt.subplots(figsize=(8, 4))
ax.bar(df["snp"], df["probability"], color="steelblue", edgecolor="black")
ax.set_ylabel("Regulatory probability")
ax.set_title("GWAS lead-SNP regulatory probabilities")
plt.xticks(rotation=45, ha="right")
plt.tight_layout()
plt.savefig("gwas_score_distribution.png", dpi=150, bbox_inches="tight")

import requests, pandas as pd
import matplotlib.pyplot as plt

BASE = "https://regulomedb.org"

def locus_profile(region, genome="GRCh38"):
    r = requests.get(f"{BASE}/regulome-search/",
                     params={"regions": region, "genome": genome, "format": "json"},
                     timeout=60)
    r.raise_for_status()
    d = r.json()
    rs = d.get("regulome_score", {})
    g = d.get("@graph", [])
    counts = pd.Series([row.get("method") for row in g]).value_counts()
    print(f"\n=== {region} | ranking={rs.get('ranking')} prob={rs.get('probability')} ===")
    print(counts.to_string())
    return counts

counts = locus_profile("chr11:5226739-5226740")  # HBB

fig, ax = plt.subplots(figsize=(8, 4))
counts.plot(kind="barh", color="seagreen", ax=ax)
ax.set_xlabel("Evidence rows in @graph")
ax.set_title("Regulatory evidence by method (HBB promoter)")
plt.tight_layout()
plt.savefig("locus_evidence_profile.png", dpi=150, bbox_inches="tight")