From ea8f531211eef8fb20465ca6e464606b4ffb12ef Mon Sep 17 00:00:00 2001
From: Matiq <kokagishi@proton.me>
Date: Sun, 6 Sep 2026 20:20:48 +0300
Subject: [PATCH] all transcripts via Ensembl REST (expand per variant,
 pan-cancer)

- vep_annotate now iterates all transcript_consequences
- timeout 15s, graceful fallback to single placeholder row
---
 annotate_clean.py | 71 ++++++++++++++++++++++++++---------------------
 1 file changed, 40 insertions(+), 31 deletions(-)

diff --git a/annotate_clean.py b/annotate_clean.py
index e3aa1f5..5a37a94 100644
--- a/annotate_clean.py
+++ b/annotate_clean.py
@@ -66,46 +66,55 @@ def vep_annotate(df, reference, gtf_path=None):
         pos1 = int(r["position"]) + 1
         ref = str(r["ref"]); alt = str(r["alt"])
         hgvs_g = f"{chrom}:g.{pos1}{ref}>{alt}"
-        gene = "intergenic"
-        hgvs_c = f"c.{pos1}{ref}>{alt}"
-        hgvs_p = f"p.(?)"
-        effect = "SNV, missense_variant (predicted)" if len(ref)==1 and len(alt)==1 else "indel"
         is_synthetic = chrom.startswith("chr") and chrom[3:].isdigit() and int(chrom[3:]) <= 3 and pos1 < 6000
         if is_synthetic:
-            gene = f"SYNTH_{chrom}"
-        else:
-            # Try Ensembl REST for real hg38 variants (one transcript, best)
-            try:
-                # Ensembl REST: GET /vep/homo_sapiens/hgvs/{hgvs_g}
-                # Use chr without prefix for Ensembl: 7:g.140453136A>T
-                hgvs_ens = f"{chrom.replace('chr','')}:g.{pos1}{ref}>{alt}"
-                url = f"https://rest.ensembl.org/vep/homo_sapiens/hgvs/{hgvs_ens}?content-type=application/json"
-                resp = requests.get(url, headers={"Content-Type": "application/json"}, timeout=5)
-                if resp.status_code == 200:
-                    j = resp.json()
-                    if j and isinstance(j, list) and j[0].get("transcript_consequences"):
-                        tc = j[0]["transcript_consequences"][0]
-                        gene = tc.get("gene_symbol") or gene
-                        hgvs_c = tc.get("hgvsc") or hgvs_c
-                        hgvs_p = tc.get("hgvsp") or hgvs_p
+            rows.append({
+                "_orig_idx": r.name,
+                "Ген": f"SYNTH_{chrom}",
+                "HGVS_c": f"c.{pos1}{ref}>{alt}",
+                "HGVS_p": "p.(?)",
+                "HGVS": f"c.{pos1}{ref}>{alt} p.(?) ({hgvs_g})",
+                "Тип варианта и эффект": "SNV, missense_variant (predicted)" if len(ref)==1 and len(alt)==1 else "indel",
+            })
+            continue
+        # Real hg38 - try Ensembl REST, expand all transcripts
+        try:
+            hgvs_ens = f"{chrom.replace('chr','')}:g.{pos1}{ref}>{alt}"
+            url = f"https://rest.ensembl.org/vep/homo_sapiens/hgvs/{hgvs_ens}?content-type=application/json"
+            resp = requests.get(url, headers={"Content-Type": "application/json", "Accept": "application/json"}, timeout=15)
+            if resp.status_code == 200:
+                j = resp.json()
+                tcs = j[0].get("transcript_consequences", []) if j and isinstance(j, list) and j[0] else []
+                if tcs:
+                    for tc in tcs:
+                        gene = tc.get("gene_symbol") or "intergenic"
+                        hgvs_c = tc.get("hgvsc") or f"c.{pos1}{ref}>{alt}"
+                        hgvs_p = tc.get("hgvsp") or "p.(?)"
                         cons = tc.get("consequence_terms", [])
-                        effect = ", ".join(cons) if cons else effect
-                        # Extract c. part from hgvsc like ENST00000288602.11:c.1799T>A -> c.1799T>A
+                        effect = ", ".join(cons) if cons else ("SNV, missense_variant (predicted)" if len(ref)==1 and len(alt)==1 else "indel")
                         if ":" in hgvs_c:
                             hgvs_c = hgvs_c.split(":")[-1]
                         if ":" in hgvs_p:
                             hgvs_p = hgvs_p.split(":")[-1]
-            except Exception:
-                pass
-        # For all-transcripts mode, we currently emit one row per variant (best transcript).
-        # With VEP cache, this would expand to N rows per variant.
+                        rows.append({
+                            "_orig_idx": r.name,
+                            "Ген": gene,
+                            "HGVS_c": hgvs_c,
+                            "HGVS_p": hgvs_p,
+                            "HGVS": f"{hgvs_c} {hgvs_p} ({hgvs_g})",
+                            "Тип варианта и эффект": effect,
+                        })
+                    continue
+        except Exception:
+            pass
+        # Fallback single row
         rows.append({
             "_orig_idx": r.name,
-            "Ген": gene,
-            "HGVS_c": hgvs_c,
-            "HGVS_p": hgvs_p,
-            "HGVS": f"{hgvs_c} {hgvs_p} ({hgvs_g})",
-            "Тип варианта и эффект": effect,
+            "Ген": "intergenic",
+            "HGVS_c": f"c.{pos1}{ref}>{alt}",
+            "HGVS_p": "p.(?)",
+            "HGVS": f"c.{pos1}{ref}>{alt} p.(?) ({hgvs_g})",
+            "Тип варианта и эффект": "SNV, missense_variant (predicted)" if len(ref)==1 and len(alt)==1 else "indel",
         })
     return pd.DataFrame(rows)