数据预处理时自动存档总荧光强度的均值与方差,以便于预测时反演荧光强度

app/api.py

@@ -124,6 +124,7 @@ class FormulationResult(BaseModel):
     all_biodist: Dict[str, float]
     # 额外预测值
     quantified_delivery: Optional[float] = None
+    unnormalized_delivery: Optional[float] = None  # 反推的原始递送值（z-score 逆变换）
     size: Optional[float] = None
     pdi_class: Optional[int] = None   # PDI 分类 (0: <0.2, 1: 0.2-0.3, 2: 0.3-0.4, 3: >0.4)
     ee_class: Optional[int] = None    # EE 分类 (0: <80%, 1: 80-90%, 2: >90%)
@@ -331,6 +332,7 @@ async def optimize_formulation(request: OptimizeRequest):
                 },
                 # 额外预测值
                 quantified_delivery=f.quantified_delivery,
+                unnormalized_delivery=f.unnormalized_delivery,
                 size=f.size,
                 pdi_class=f.pdi_class,
                 ee_class=f.ee_class,

app/app.py

@@ -227,6 +227,8 @@ def format_results_dataframe(results: dict, smiles_label: str = None) -> pd.Data
         # 添加额外预测值
         if f.get("quantified_delivery") is not None:
             row["量化递送"] = f"{f['quantified_delivery']:.4f}"
+        if f.get("unnormalized_delivery") is not None:
+            row["总荧光强度"] = f"{f['unnormalized_delivery']:.4f}"
         if f.get("size") is not None:
             row["粒径(nm)"] = f"{f['size']:.1f}"
         if f.get("pdi_class") is not None:

app/delivery_zscore_stats.json

@@ -0,0 +1,14 @@
+{
+  "intramuscular": {
+    "mean": 0.7281303554081238,
+    "std": 0.7006554090148486,
+    "qd_min": -1.0387570720282182,
+    "qd_max": 4.73706835052163
+  },
+  "intravenous": {
+    "mean": 0.29940387649347033,
+    "std": 0.37474351840219583,
+    "qd_min": -0.7985592911689305,
+    "qd_max": 4.497814051056962
+  }
+}

app/optimize.py

@@ -8,6 +8,7 @@
 """
 
 import itertools
+import json
 from pathlib import Path
 from typing import Dict, List, Optional, Tuple
 from dataclasses import dataclass, field
@@ -124,11 +125,25 @@ HELPER_LIPID_OPTIONS = ["DOPE", "DSPC"]
 # Route of administration 选项
 ROUTE_OPTIONS = ["intravenous", "intramuscular"]
 
-# quantified_delivery 归一化常量（按给药途径）
-DELIVERY_NORM = {
-    "intravenous": {"min": -0.798559291, "max": 4.497814051056962},
-    "intramuscular": {"min": -0.794912427, "max": 10.220042980012716},
-}
+# delivery 统计量（由 preprocess_internal.py 生成）
+# 包含: mean/std（z-score 逆变换）、qd_min/qd_max（评分归一化）
+_DELIVERY_STATS_PATH = Path(__file__).resolve().parent / "delivery_zscore_stats.json"
+if _DELIVERY_STATS_PATH.exists():
+    with open(_DELIVERY_STATS_PATH) as _f:
+        DELIVERY_ZSCORE_STATS: Dict[str, Dict[str, float]] = json.load(_f)
+    logger.info(f"Loaded delivery stats from {_DELIVERY_STATS_PATH}")
+else:
+    DELIVERY_ZSCORE_STATS = {}
+    logger.warning(f"delivery_zscore_stats.json not found at {_DELIVERY_STATS_PATH}, "
+                   "run 'make preprocess' to generate it")
+
+# quantified_delivery 归一化常量（从统计量中提取 qd_min/qd_max，用于评分归一化到 [0,1]）
+DELIVERY_NORM: Dict[str, Dict[str, float]] = {}
+for _route, _stats in DELIVERY_ZSCORE_STATS.items():
+    if "qd_min" in _stats and "qd_max" in _stats:
+        DELIVERY_NORM[_route] = {"min": _stats["qd_min"], "max": _stats["qd_max"]}
+if not DELIVERY_NORM:
+    logger.warning("DELIVERY_NORM is empty — scoring normalization for delivery will be disabled")
 
 
 @dataclass
@@ -282,6 +297,7 @@ class Formulation:
     biodist_predictions: Dict[str, float] = field(default_factory=dict)
     # 额外预测值
     quantified_delivery: Optional[float] = None
+    unnormalized_delivery: Optional[float] = None  # 反推的原始递送值（z-score 逆变换）
     size: Optional[float] = None
     pdi_class: Optional[int] = None   # PDI 分类 (0-3)
     ee_class: Optional[int] = None    # EE 分类 (0-2)
@@ -587,6 +603,16 @@ def predict_all(
     df["pred_pdi_class"] = pdi_preds
     df["pred_ee_class"] = ee_preds
     df["pred_toxic_class"] = toxic_preds
+
+    # 反推 unnormalized_delivery: value = z-score * std + mean
+    df["pred_unnorm_delivery"] = np.nan
+    if DELIVERY_ZSCORE_STATS:
+        for route_name, stats in DELIVERY_ZSCORE_STATS.items():
+            mask = df["_route"] == route_name
+            if mask.any():
+                df.loc[mask, "pred_unnorm_delivery"] = (
+                    delivery_preds[mask.values] * stats["std"] + stats["mean"]
+                )
     
     return df
 
@@ -645,6 +671,9 @@ def select_top_k(
         
         if key not in seen:
             seen.add(key)
+            unnorm_val = row.get("pred_unnorm_delivery")
+            unnorm_delivery = float(unnorm_val) if pd.notna(unnorm_val) else None
+
             formulation = Formulation(
                 cationic_lipid_to_mrna_ratio=row["Cationic_Lipid_to_mRNA_weight_ratio"],
                 cationic_lipid_mol_ratio=row["Cationic_Lipid_Mol_Ratio"],
@@ -658,6 +687,7 @@ def select_top_k(
                 },
                 # 额外预测值
                 quantified_delivery=row.get("pred_delivery"),
+                unnormalized_delivery=unnorm_delivery,
                 size=row.get("pred_size"),
                 pdi_class=int(row.get("pred_pdi_class")) if row.get("pred_pdi_class") is not None else None,
                 ee_class=int(row.get("pred_ee_class")) if row.get("pred_ee_class") is not None else None,

scripts/preprocess_internal.py

@@ -1,5 +1,6 @@
 """数据清洗脚本：修正原始数据中的问题"""
 
+import json
 from pathlib import Path
 
 import numpy as np
@@ -9,6 +10,7 @@ from loguru import logger
 
 from lnp_ml.config import RAW_DATA_DIR, INTERIM_DATA_DIR
 
+APP_DIR = Path(__file__).resolve().parents[1] / "app"
 
 app = typer.Typer()
 
@@ -24,6 +26,7 @@ def main(
     修正内容：
     1. 按给药途径分组进行 z-score 标准化
     2. 对 size 列取 log
+    3. 将 z-score 的 mean/std 保存到 app/ 供推理时反推
     """
     logger.info(f"Loading data from {input_path}")
     df = pd.read_excel(input_path, header=2)
@@ -32,11 +35,34 @@ def main(
     # 分别对肌肉注射组和静脉注射组重新进行 z-score 标准化
     logger.info("Z-score normalizing delivery by Route_of_administration...")
     df["unnormalized_delivery"] = pd.to_numeric(df["unnormalized_delivery"], errors="coerce")
+
+    # 计算并保存 per-route 统计量，用于推理时反推和评分归一化
+    zscore_stats = {}
+    for route, group in df.groupby("Route_of_administration"):
+        vals = group["unnormalized_delivery"].dropna()
+        if len(vals) > 1:
+            zscore_stats[route] = {"mean": float(vals.mean()), "std": float(vals.std())}
+            logger.info(f"  {route}: mean={vals.mean():.6f}, std={vals.std():.6f}, n={len(vals)}")
+
     df["quantified_delivery"] = (
         df.groupby("Route_of_administration")["unnormalized_delivery"]
           .transform(lambda x: (x - x.mean()) / x.std())
     )
 
+    # 补充 quantified_delivery 的 per-route min/max（用于评分时归一化到 [0,1]）
+    for route, group in df.groupby("Route_of_administration"):
+        qd = group["quantified_delivery"].dropna()
+        if len(qd) > 0 and route in zscore_stats:
+            zscore_stats[route]["qd_min"] = float(qd.min())
+            zscore_stats[route]["qd_max"] = float(qd.max())
+            logger.info(f"  {route}: qd_min={qd.min():.6f}, qd_max={qd.max():.6f}")
+
+    stats_path = APP_DIR / "delivery_zscore_stats.json"
+    stats_path.parent.mkdir(parents=True, exist_ok=True)
+    with open(stats_path, "w") as f:
+        json.dump(zscore_stats, f, indent=2)
+    logger.success(f"Saved delivery stats to {stats_path}")
+
     # 对 size 列取 log
     logger.info("Log-transforming size column...")
     df["size"] = pd.to_numeric(df["size"], errors="coerce")