| """Data classes for an CasMD report + JSON loader.""" |
| from __future__ import annotations |
| import json |
| from dataclasses import dataclass |
| from pathlib import Path |
|
|
|
|
| @dataclass(frozen=True) |
| class PredictionSummary: |
| backend: str |
| model_id: int |
| iptm: float | None |
| ptm: float | None |
| plddt_mean: float | None |
|
|
|
|
| @dataclass(frozen=True) |
| class AnalysisSummary: |
| n_frames: int |
| equil_skip: int |
| protein_rmsd_equil_mean_A: float | None |
| protein_rmsd_max_A: float | None |
| protein_rmsf_mean_A: float | None |
| protein_rg_equil_mean_A: float | None |
|
|
|
|
| @dataclass(frozen=True) |
| class ReportData: |
| job_name: str |
| production_ns: float |
| prediction: PredictionSummary | None |
| analysis: AnalysisSummary |
| figures: dict[str, Path] |
| interpretation: str |
|
|
|
|
| @dataclass |
| class AnalysisResult: |
| """Extended analysis result for comparison workflows. |
| |
| Mirrors the scalar fields of AnalysisSummary but omits protein_rmsd_max_A |
| (not needed for comparison) and adds optional per-frame series fields so |
| that statistical tests (Welch's t-test, KS) can operate on raw data. |
| """ |
| n_frames: int |
| equil_skip: int |
| protein_rmsd_equil_mean_A: float | None |
| protein_rmsf_mean_A: float | None |
| protein_rg_equil_mean_A: float | None |
| |
| rmsd_series: list[float] | None = None |
| rmsf_series: list[float] | None = None |
| rg_series: list[float] | None = None |
|
|
|
|
| def load_analysis_json(path: Path) -> AnalysisSummary: |
| """Parse the `results.json` produced by Plan 3's templated analyze.py.""" |
| path = Path(path) |
| data = json.loads(path.read_text()) |
| rmsd = data.get("protein_rmsd_A", {}) |
| rg = data.get("protein_rg_A", {}) |
| return AnalysisSummary( |
| n_frames=int(data.get("n_frames", 0)), |
| equil_skip=int(data.get("equil_skip", 0)), |
| protein_rmsd_equil_mean_A=_maybe_float(rmsd.get("equilibrated_mean")), |
| protein_rmsd_max_A=_maybe_float(rmsd.get("max")), |
| protein_rmsf_mean_A=_maybe_float(data.get("protein_rmsf_A_mean")), |
| protein_rg_equil_mean_A=_maybe_float(rg.get("equilibrated_mean")), |
| ) |
|
|
|
|
| def _maybe_float(v) -> float | None: |
| if v is None: |
| return None |
| try: |
| return float(v) |
| except (TypeError, ValueError): |
| return None |
|
|
