Update src/validation/external_benchmark.py

src/validation/external_benchmark.py

@@ -15,16 +15,7 @@ YOR1,NaCl,resistant,0.06,HIP-HOP_demo
 ATM1,NaCl,resistant,0.10,HIP-HOP_demo
 """)
 
-# === Section 6.2 — External Benchmark (HIP-HOP / MoAmap / SGD) ===
-# Looks for any of these files (CSV/TSV). Put them anywhere under data/external/.
-#   HIP-HOP fitness:   gene, condition/compound, effect(or fitness/logFC), pval/FDR (optional)
-#   MoAmap signatures: gene, compound, correlation/effect, pval/FDR (optional)
-#   SGD phenotypes:    gene, phenotype, condition (free text), direction (e.g., 'sensitive','resistant')
-#
-# Output:
-#   results/validation_external_matrix.csv      (#evidence per gene×stress, signed)
-#   results/validation_external_concordance.csv (direction vs our ATE)
-#   results/validation_external_heatmap.png
+
 
 import os, re, json, glob, numpy as np, pandas as pd, matplotlib.pyplot as plt, seaborn as sns
 from pathlib import Path
@@ -32,7 +23,6 @@ from pathlib import Path
 RES = Path("results"); RES.mkdir(exist_ok=True, parents=True)
 EXT = Path("data/external")
 
-# ---------- helpers ----------
 def _read_any(path):
     if path.endswith(".tsv") or path.endswith(".tab"):
         return pd.read_csv(path, sep="\t", dtype=str, low_memory=False)
@@ -44,10 +34,9 @@ def _coerce_float(s):
 
 def _norm_gene(x):
     x = (str(x) or "").strip()
-    x = re.sub(r"_expr$", "", x)          # our columns like PDR5_expr
+    x = re.sub(r"_expr$", "", x)          
     x = x.upper()
-    # common yeast ORF → symbol hints (very light-weight; add more if you like)
-    x = re.sub(r"^Y[A-P][LR][0-9]{3}[CW](-[A-Z])?$", x, x)  # keep ORF if already ORF
+    x = re.sub(r"^Y[A-P][LR][0-9]{3}[CW](-[A-Z])?$", x, x)  
     return x
 
 def _stress_from_text(t):
@@ -63,7 +52,6 @@ def _sign_from_effect(val, dir_text=None):
         d = str(dir_text).lower()
         if any(k in d for k in ["resist", "increased tolerance", "gain"]): return +1
         if any(k in d for k in ["sensit", "hypersens", "loss"]): return -1
-    # fall back to numeric sign
     try:
         v = float(val)
         if np.isnan(v): return 0
@@ -71,10 +59,8 @@ def _sign_from_effect(val, dir_text=None):
     except:
         return 0
 
-# ---------- load our ATEs ----------
 snap = json.load(open(RES/"causal_section3_snapshot.json"))
 ATE_table = snap.get("ATE_table") or snap.get("stress_ate") or {}
-# flatten to transporter→stress→ATE
 flat = []
 for k,v in ATE_table.items():
     g = _norm_gene(k)
@@ -89,7 +75,6 @@ if df_ate.empty:
 # anchors for quick viewing later
 anchors = ["PDR5","SNQ2","YOR1","ATM1"]
 
-# ---------- scan external sources ----------
 files = []
 for pat in ["**/*hip*hop*.csv","**/*hip*hop*.tsv",
             "**/*moa*map*.csv","**/*moa*map*.tsv",
@@ -128,30 +113,26 @@ for f in sorted(set(files)):
 
 ext = pd.DataFrame(ext_rows)
 if ext.empty:
-    print("⚠️ No usable external rows found under", EXT.resolve())
+    print(" No usable external rows found under", EXT.resolve())
 else:
     print(f"Loaded external evidence rows: {len(ext)} from {ext['source'].nunique()} files")
 
-# ---------- aggregate external evidence ----------
 if ext.empty:
-    # still write empty shells, so the paper pipeline doesn't break
     pd.DataFrame(columns=["gene","stress","evidence_n","evidence_balance"]).to_csv(RES/"validation_external_matrix.csv", index=False)
     pd.DataFrame(columns=["gene","stress","ATE","ext_consensus","concordant"]).to_csv(RES/"validation_external_concordance.csv", index=False)
 else:
     agg = (ext
            .groupby(["gene","stress"])["sign"]
-           .agg(evidence_n="count", evidence_balance="sum")  # + counts as resistance, - as sensitivity
+           .agg(evidence_n="count", evidence_balance="sum") 
            .reset_index())
     agg.to_csv(RES/"validation_external_matrix.csv", index=False)
 
-    # merge with our ATEs and compute concordance of direction
     M = df_ate.merge(agg, on=["gene","stress"], how="left")
     M["ext_consensus"] = np.sign(M["evidence_balance"].fillna(0))
     M["ate_sign"] = np.sign(M["ATE"])
     M["concordant"] = (M["ext_consensus"] != 0) & (M["ate_sign"] == M["ext_consensus"])
     M.to_csv(RES/"validation_external_concordance.csv", index=False)
 
-    # quick heatmap of external consensus vs our ATE (anchors first if present)
     order_genes = [g for g in anchors if g in set(M["gene"])] + [g for g in M["gene"].unique() if g not in anchors]
     pt = M.pivot_table(index="gene", columns="stress", values="ext_consensus", aggfunc="first").reindex(order_genes)
     plt.figure(figsize=(7, max(3, 0.35*len(pt))))
@@ -159,7 +140,6 @@ else:
     plt.title("External benchmark — consensus sign (HIP-HOP/MoAmap/SGD)")
     plt.tight_layout(); plt.savefig(RES/"validation_external_heatmap.png", dpi=300); plt.show()
 
-    # short summary for the anchors
     summ = (M[M["gene"].isin(anchors)]
             .groupby("gene")[["concordant"]]
             .mean().rename(columns={"concordant":"concordance_rate"}))
@@ -171,13 +151,11 @@ print("Saved:",
       RES/"validation_external_concordance.csv",
       RES/"validation_external_heatmap.png")
 
-# === External benchmark heatmap (robust) ===
 import numpy as np, pandas as pd, seaborn as sns, matplotlib.pyplot as plt, pathlib as p
 
 RES = p.Path("results")
 concord = pd.read_csv(RES/"validation_external_concordance.csv")
 
-# Find/derive the external sign column
 sign_col = None
 for c in concord.columns:
     if "sign" in c.lower() and c.lower() != "atlas_sign":
@@ -185,12 +163,10 @@ for c in concord.columns:
         break
 
 if sign_col is None:
-    # Fall back to the wide matrix and create a long "sign" table
-    mat = pd.read_csv(RES/"validation_external_matrix.csv")  # columns: gene, stress, value (or one col per stress)
+    mat = pd.read_csv(RES/"validation_external_matrix.csv")  
     if {"gene","stress","value"}.issubset(mat.columns):
         ext_long = mat.rename(columns={"value":"external_consensus_sign"})
     else:
-        # wide → long
         long_rows=[]
         wide_stresses = [c for c in mat.columns if c not in ["gene","Gene","GENE"]]
         gcol = "gene" if "gene" in mat.columns else ("Gene" if "Gene" in mat.columns else "GENE")
@@ -200,15 +176,13 @@ if sign_col is None:
                 long_rows.append({"gene": g, "stress": s, "external_consensus_sign": getattr(r, s)})
         ext_long = pd.DataFrame(long_rows)
 else:
-    # Use the column we found in concordance
     ext_long = concord[["gene","stress",sign_col]].rename(columns={sign_col:"external_consensus_sign"})
 
-# Keep only non-zero/finite entries
 ext_long["external_consensus_sign"] = pd.to_numeric(ext_long["external_consensus_sign"], errors="coerce").fillna(0.0)
 ext_plot = ext_long[np.abs(ext_long["external_consensus_sign"]) > 0].copy()
 
 if ext_plot.empty:
-    print("⚠️ No non-zero benchmark entries to plot. Check your CSV contents.")
+    print(" No non-zero benchmark entries to plot. Check your CSV contents.")
 else:
     pv = ext_plot.pivot(index="gene", columns="stress", values="external_consensus_sign")
     plt.figure(figsize=(6, max(3, 0.4*pv.shape[0])))
@@ -218,4 +192,4 @@ else:
     plt.tight_layout()
     out = RES/"validation_external_subset_heatmap.png"
     plt.savefig(out, dpi=300); plt.show()
-    print("✅ Saved:", out)
+    print(" Saved:", out)