91 lines
3.2 KiB
Python
91 lines
3.2 KiB
Python
|
from __future__ import annotations
|
||
|
|
|
||
|
|
import json
|
||
|
|
import subprocess
|
||
|
|
import sys
|
||
|
|
from pathlib import Path
|
||
|
|
|
||
|
|
ROOT = Path(__file__).resolve().parents[2]
|
||
|
|
|
||
|
|
|
||
|
|
def test_msa_region_plot_inherits_scope_from_source_manifest(tmp_path):
|
||
|
|
workspace = tmp_path / "job_msa_scope"
|
||
|
|
final_dir = workspace / "final"
|
||
|
|
final_dir.mkdir(parents=True)
|
||
|
|
prefix = "Target.SPL.target-Lc.ref-At.subfamily_prediction"
|
||
|
|
|
||
|
|
msa = final_dir / f"{prefix}.pep.msa.fa"
|
||
|
|
msa.write_text(
|
||
|
|
">LcGene1\nACDEFGHIKL\n"
|
||
|
|
">LcGene2\nACDEYGHIKL\n"
|
||
|
|
">AtGene1\nACDEFGHIKL\n",
|
||
|
|
encoding="utf-8",
|
||
|
|
)
|
||
|
|
|
||
|
|
source_manifest = final_dir / f"{prefix}.phylogeny.manifest.json"
|
||
|
|
source_manifest.write_text(
|
||
|
|
json.dumps({
|
||
|
|
"manifest_version": "1.0",
|
||
|
|
"stage": "phylogeny",
|
||
|
|
"status": "success",
|
||
|
|
"prefix": "Lc.SPL",
|
||
|
|
"output_prefix": prefix,
|
||
|
|
"scope_id": "SPL.target-Lc.ref-At.target_plus_reference",
|
||
|
|
"scope_mode": "target_plus_reference",
|
||
|
|
"target_species": ["Lc"],
|
||
|
|
"reference_species": ["At"],
|
||
|
|
"comparative_species": [],
|
||
|
|
"selected_species": ["Lc", "At"],
|
||
|
|
"analysis_route": "subfamily_prediction",
|
||
|
|
"member_set_id": "SPL.target-Lc.ref-At.target_plus_reference:subfamily_prediction:all_members",
|
||
|
|
"artifacts": {"msa": {"full_path": str(msa), "type": "fasta"}},
|
||
|
|
}),
|
||
|
|
encoding="utf-8",
|
||
|
|
)
|
||
|
|
|
||
|
|
fake_r = tmp_path / "fake_msa_plot.sh"
|
||
|
|
fake_r.write_text(
|
||
|
|
"#!/usr/bin/env bash\n"
|
||
|
|
"outbase=\"$2\"\n"
|
||
|
|
"printf fake > \"${outbase}_msa.png\"\n"
|
||
|
|
"printf fake > \"${outbase}_msa.pdf\"\n",
|
||
|
|
encoding="utf-8",
|
||
|
|
)
|
||
|
|
fake_r.chmod(0o755)
|
||
|
|
|
||
|
|
proc = subprocess.run(
|
||
|
|
[
|
||
|
|
sys.executable,
|
||
|
|
str(ROOT / "apps/cli/run_msa_region_plot.py"),
|
||
|
|
"--workspace", str(workspace),
|
||
|
|
"--prefix", prefix,
|
||
|
|
"--msa-file", str(msa),
|
||
|
|
"--auto-conserved",
|
||
|
|
"--window-len", "4",
|
||
|
|
"--min-nongap", "0",
|
||
|
|
"--source-manifest", str(source_manifest),
|
||
|
|
"--rscript", "/bin/bash",
|
||
|
|
"--r-script", str(fake_r),
|
||
|
|
"--force",
|
||
|
|
],
|
||
|
|
cwd=str(ROOT),
|
||
|
|
capture_output=True,
|
||
|
|
text=True,
|
||
|
|
check=False,
|
||
|
|
)
|
||
|
|
|
||
|
|
assert proc.returncode == 0, proc.stderr or proc.stdout
|
||
|
|
manifests = list(final_dir.glob(f"{prefix}.msa_conserved_region_*.manifest.json"))
|
||
|
|
assert len(manifests) == 1
|
||
|
|
manifest = json.loads(manifests[0].read_text(encoding="utf-8"))
|
||
|
|
|
||
|
|
assert manifest["scope_id"] == "SPL.target-Lc.ref-At.target_plus_reference"
|
||
|
|
assert manifest["target_species"] == ["Lc"]
|
||
|
|
assert manifest["reference_species"] == ["At"]
|
||
|
|
assert manifest["analysis_route"] == "subfamily_prediction"
|
||
|
|
assert manifest["member_set_id"] == "SPL.target-Lc.ref-At.target_plus_reference:subfamily_prediction:all_members"
|
||
|
|
assert manifest["stage_context"]["scope_id"] == "SPL.target-Lc.ref-At.target_plus_reference"
|
||
|
|
assert manifest["parameters"]["source_manifest"] == str(source_manifest.resolve())
|
||
|
|
assert manifest["source_manifest"] == str(source_manifest.resolve())
|
||
|
|
assert any(path.endswith("_msa.png") for path in manifest["artifacts"]["output_files"])
|