kemal
5463f489a3
ci / api (push) Failing after 10s
ci / terraform (push) Failing after 11s
ci / web (push) Failing after 35s
ci / pipeline (push) Failing after 2m29s
ci / images (api) (push) Skipped
ci / images (ml) (push) Skipped
ci / images (pipeline) (push) Skipped
ci / images (web) (push) Skipped
End-to-end variant interpretation platform for rare genetic disease research: SvelteKit UI, FastAPI + PostgreSQL API, Nextflow/Ensembl VEP pipeline, LightGBM pathogenicity scoring with MLflow, K8s/ArgoCD/GCP infrastructure. Public test data only; no clinical claims.
56 lines
2.1 KiB
Python
56 lines
2.1 KiB
Python
"""Train a pathogenicity classifier on ClinVar labels (Pathogenic/Likely pathogenic vs Benign/Likely benign).
|
|
|
|
Label leakage warning: CLIN_SIG must never be a feature. This is a learning exercise, not a clinical model.
|
|
Usage: python -m rarelens_ml.train --tsv results/clinvar.vep.tsv
|
|
"""
|
|
import argparse
|
|
|
|
import lightgbm as lgb
|
|
import mlflow
|
|
import mlflow.lightgbm
|
|
import pandas as pd
|
|
from sklearn.metrics import average_precision_score, roc_auc_score
|
|
from sklearn.model_selection import train_test_split
|
|
|
|
from rarelens_ml.features import build
|
|
|
|
POS = {"Pathogenic", "Likely_pathogenic", "Pathogenic/Likely_pathogenic"}
|
|
NEG = {"Benign", "Likely_benign", "Benign/Likely_benign"}
|
|
|
|
|
|
def load(tsv: str) -> tuple[pd.DataFrame, pd.Series]:
|
|
df = pd.read_csv(tsv, sep="\t", comment="#", header=None, dtype=str)
|
|
with open(tsv) as fh:
|
|
df.columns = next(l for l in fh if l.startswith("#Uploaded")).lstrip("#").rstrip().split("\t")
|
|
df = df.rename(columns={"IMPACT": "impact", "Consequence": "consequence", "gnomADe_AF": "gnomad_af",
|
|
"CADD_PHRED": "cadd_phred"})
|
|
y = df["CLIN_SIG"].map(lambda s: 1 if s in POS else 0 if s in NEG else None)
|
|
keep = y.notna()
|
|
return build(df[keep]), y[keep].astype(int)
|
|
|
|
|
|
def main() -> None:
|
|
p = argparse.ArgumentParser()
|
|
p.add_argument("--tsv", required=True)
|
|
p.add_argument("--register", action="store_true")
|
|
a = p.parse_args()
|
|
|
|
X, y = load(a.tsv)
|
|
Xtr, Xte, ytr, yte = train_test_split(X, y, test_size=0.2, stratify=y, random_state=42)
|
|
|
|
mlflow.set_experiment("rarelens-pathogenicity")
|
|
with mlflow.start_run():
|
|
params = {"n_estimators": 400, "learning_rate": 0.05, "num_leaves": 31, "class_weight": "balanced"}
|
|
mlflow.log_params(params)
|
|
model = lgb.LGBMClassifier(**params).fit(Xtr, ytr)
|
|
proba = model.predict_proba(Xte)[:, 1]
|
|
mlflow.log_metrics({"auroc": roc_auc_score(yte, proba), "auprc": average_precision_score(yte, proba)})
|
|
mlflow.lightgbm.log_model(
|
|
model, "model",
|
|
registered_model_name="rarelens-pathogenicity" if a.register else None,
|
|
)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|