File size: 3,789 Bytes
db32e07 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 | #!/usr/bin/env python
"""
Prepare JSONL for Spectra-Reason-GCD: peaks (list of [mz, int]) + smiles.
Input: TSV with columns mzs, intensities, smiles (e.g. MassSpecGym) or MGF + metadata.
"""
from __future__ import annotations
import argparse
import json
import sys
from pathlib import Path
ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
sys.path.insert(0, str(ROOT))
def parse_spec_array(s: str) -> list[float]:
if isinstance(s, (list, tuple)):
return [float(x) for x in s]
return [float(x) for x in str(s).split(",") if x.strip()]
def parse_args() -> argparse.Namespace:
p = argparse.ArgumentParser(description="Prepare peaks+smiles JSONL for CoT training.")
p.add_argument("--tsv", default=None, help="TSV with mzs, intensities, smiles columns")
p.add_argument("--mgf", default=None, help="MGF file (requires smiles in TITLE or separate mapping)")
p.add_argument("--output-jsonl", required=True)
p.add_argument("--max-peaks", type=int, default=60)
p.add_argument("--fold", default=None, help="If TSV has 'fold', filter to this (e.g. train)")
return p.parse_args()
def main() -> None:
args = parse_args()
out_path = Path(args.output_jsonl)
out_path.parent.mkdir(parents=True, exist_ok=True)
records = []
if args.tsv:
import csv
with open(args.tsv, "r", encoding="utf-8") as f:
reader = csv.DictReader(f, delimiter="\t")
rows = list(reader)
if not rows or "mzs" not in rows[0] or "intensities" not in rows[0] or "smiles" not in rows[0]:
raise ValueError("TSV must have mzs, intensities, smiles columns")
for row in rows:
if args.fold and "fold" in row and row.get("fold") != args.fold:
continue
mzs = parse_spec_array(row["mzs"])
intens = parse_spec_array(row["intensities"])
peaks = [[m, i] for m, i in zip(mzs, intens)]
peaks = sorted(peaks, key=lambda x: x[1], reverse=True)[: args.max_peaks]
rec = {
"peaks": peaks,
"smiles": str(row["smiles"]).strip(),
}
if "formula" in row and row["formula"]:
rec["formula"] = str(row["formula"]).strip()
if "precursor_mz" in row and row["precursor_mz"]:
try:
rec["precursor_mz"] = float(row["precursor_mz"])
except (ValueError, TypeError):
pass
if "precursor_formula" in row and row["precursor_formula"] and "formula" not in rec:
rec["formula"] = str(row["precursor_formula"]).strip()
records.append(rec)
elif args.mgf:
try:
from pyteomics import mgf
except ImportError:
raise ImportError("pyteomics required for MGF: pip install pyteomics")
with mgf.MGF(args.mgf) as reader:
for spec in reader:
mz = spec.get("m/z array", [])
iarr = spec.get("intensity array", [])
peaks = [[float(m), float(i)] for m, i in zip(mz, iarr)]
peaks = sorted(peaks, key=lambda x: x[1], reverse=True)[: args.max_peaks]
smiles = spec.get("params", {}).get("SMILES", spec.get("params", {}).get("smiles", ""))
if not smiles and "TITLE" in spec.get("params", {}):
smiles = spec["params"]["TITLE"]
records.append({"peaks": peaks, "smiles": str(smiles).strip()})
else:
raise ValueError("Provide --tsv or --mgf")
with open(out_path, "w") as f:
for r in records:
f.write(json.dumps(r) + "\n")
print(f"Wrote {len(records)} records to {out_path}")
if __name__ == "__main__":
main()
|