Skip to content

174 feat add mgf file support for mztab data loader #178

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
JemmaLDaniel wants to merge 5 commits into from
Closed

174 feat add mgf file support for mztab data loader #178

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
fb27d99
feat: support mgf file loading and auto-addition of spectrum ID for M…
JemmaLDaniel Apr 4, 2026
abcdd5b
test: add tests for auto-addition of spectrum ID and mgf file loading…
JemmaLDaniel Apr 4, 2026
72f017e
chore: default to not adding spectrum_id columns
JemmaLDaniel Jun 15, 2026
cb3b7b5
docs: update data loader documentation to clarify matching protocol
JemmaLDaniel Jun 15, 2026
ce03259
Merge branch 'main' into 174-feat-add-mgf-file-support-for-mztab-data…
JemmaLDaniel Jun 15, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
120 changes: 110 additions & 10 deletions tests/datasets/test_data_loaders.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1150,44 +1150,144 @@ def test_only_unimod_notation_needed_in_invalid_list(
# _load_spectrum_data
# ------------------------------------------------------------------

def test_load_spectrum_data_raises_for_unsupported_extension(self, tmp_path):
def test_load_spectrum_data_raises_for_unsupported_extension(
self, loader, tmp_path
):
path = tmp_path / "data.tsv"
path.touch()
with pytest.raises(ValueError, match="Unsupported file format"):
MZTabDatasetLoader._load_spectrum_data(path)
loader._load_spectrum_data(path)

def test_load_spectrum_data_reads_parquet(self, tmp_path):
def test_load_spectrum_data_reads_parquet(self, loader, tmp_path):
df = pl.DataFrame({"charge": [2], "mz_array": [[100.0]]})
path = tmp_path / "data.parquet"
df.write_parquet(path)

result_df, _ = MZTabDatasetLoader._load_spectrum_data(path)
result_df, _ = loader._load_spectrum_data(path)
assert "charge" in result_df.columns

def test_load_spectrum_data_reads_ipc(self, tmp_path):
def test_load_spectrum_data_reads_ipc(self, loader, tmp_path):
df = pl.DataFrame({"charge": [2], "mz_array": [[100.0]]})
path = tmp_path / "data.ipc"
df.write_ipc(path)

result_df, _ = MZTabDatasetLoader._load_spectrum_data(path)
result_df, _ = loader._load_spectrum_data(path)
assert "charge" in result_df.columns

def test_load_spectrum_data_detects_labels_when_sequence_present(self, tmp_path):
def test_load_spectrum_data_detects_labels_when_sequence_present(
self, loader, tmp_path
):
df = pl.DataFrame({"sequence": ["PEPTIDE"], "charge": [2]})
path = tmp_path / "data.parquet"
df.write_parquet(path)

_, has_labels = MZTabDatasetLoader._load_spectrum_data(path)
_, has_labels = loader._load_spectrum_data(path)
assert has_labels is True

def test_load_spectrum_data_no_labels_when_sequence_absent(self, tmp_path):
def test_load_spectrum_data_no_labels_when_sequence_absent(self, loader, tmp_path):
df = pl.DataFrame({"charge": [2]})
path = tmp_path / "data.parquet"
df.write_parquet(path)

_, has_labels = MZTabDatasetLoader._load_spectrum_data(path)
_, has_labels = loader._load_spectrum_data(path)
assert has_labels is False

def test_load_spectrum_data_reads_mgf(self, loader, tmp_path):
mgf_path = tmp_path / "spectra.mgf"
mgf_path.write_text(
"BEGIN IONS\n"
"PEPMASS=500.0\n"
"CHARGE=2+\n"
"RTINSECONDS=100.0\n"
"100.0 1.0\n"
"END IONS\n",
encoding="utf-8",
)
result_df, has_labels = loader._load_spectrum_data(mgf_path)
assert "mz_array" in result_df.columns
assert "intensity_array" in result_df.columns
assert "precursor_mz" in result_df.columns
assert "precursor_charge" in result_df.columns
assert has_labels is False

def test_load_spectrum_data_mgf_always_adds_index_cols(self, tmp_path):
"""MGF inputs get experiment_name and spectrum_id regardless of add_index_cols."""
loader = MZTabDatasetLoader(
residue_masses=_FULL_RESIDUE_MASSES,
residue_remapping=_STANDARD_REMAPPING,
add_index_cols=False,
)
mgf_path = tmp_path / "spectra.mgf"
mgf_path.write_text(
"BEGIN IONS\n" "PEPMASS=500.0\n" "CHARGE=2+\n" "100.0 1.0\n" "END IONS\n",
encoding="utf-8",
)
result_df, _ = loader._load_spectrum_data(mgf_path)
assert "experiment_name" in result_df.columns
assert "spectrum_id" in result_df.columns
assert result_df["spectrum_id"][0] == "spectra:0"

def test_load_spectrum_data_mgf_has_labels(self, loader, tmp_path):
mgf_path = tmp_path / "labeled.mgf"
mgf_path.write_text(
"BEGIN IONS\n"
"PEPMASS=500.0\n"
"CHARGE=2+\n"
"SEQ=PEPTIDE\n"
"100.0 1.0\n"
"END IONS\n",
encoding="utf-8",
)
_, has_labels = loader._load_spectrum_data(mgf_path)
assert has_labels is True

def test_load_spectrum_data_mgf_no_labels(self, loader, tmp_path):
mgf_path = tmp_path / "unlabeled.mgf"
mgf_path.write_text(
"BEGIN IONS\n" "PEPMASS=500.0\n" "CHARGE=2+\n" "100.0 1.0\n" "END IONS\n",
encoding="utf-8",
)
_, has_labels = loader._load_spectrum_data(mgf_path)
assert has_labels is False

def test_load_spectrum_data_parquet_adds_index_cols_when_enabled(self, tmp_path):
loader = MZTabDatasetLoader(
residue_masses=_FULL_RESIDUE_MASSES,
residue_remapping=_STANDARD_REMAPPING,
add_index_cols=True,
)
df = pl.DataFrame({"charge": [2], "mz_array": [[100.0]]})
path = tmp_path / "spec.parquet"
df.write_parquet(path)
result_df, _ = loader._load_spectrum_data(path)
assert "experiment_name" in result_df.columns
assert "spectrum_id" in result_df.columns
assert result_df["spectrum_id"][0] == "spec:0"

def test_load_spectrum_data_ipc_adds_index_cols_when_enabled(self, tmp_path):
loader = MZTabDatasetLoader(
residue_masses=_FULL_RESIDUE_MASSES,
residue_remapping=_STANDARD_REMAPPING,
add_index_cols=True,
)
df = pl.DataFrame({"charge": [2], "mz_array": [[100.0]]})
path = tmp_path / "spec.ipc"
df.write_ipc(path)
result_df, _ = loader._load_spectrum_data(path)
assert "experiment_name" in result_df.columns
assert "spectrum_id" in result_df.columns
assert result_df["spectrum_id"][0] == "spec:0"

def test_load_spectrum_data_parquet_no_index_cols_by_default(
self, loader, tmp_path
):
df = pl.DataFrame({"charge": [2], "mz_array": [[100.0]]})
path = tmp_path / "data.parquet"
df.write_parquet(path)
result_df, _ = loader._load_spectrum_data(path)
assert "experiment_name" not in result_df.columns
assert "spectrum_id" not in result_df.columns

# ------------------------------------------------------------------
# _load_dataset
# ------------------------------------------------------------------
Expand Down
6 changes: 6 additions & 0 deletions winnow/configs/data_loader/mztab.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,12 @@ residue_masses: ${residue_masses}
# Defaults to true.
load_beams: true

# When true, add experiment_name + spectrum_id to spectrum data. This maps input spectra
# to search engine output predictions (e.g. Casanovo, InstaNovo). MGF files always receive
# these columns regardless of this setting. Set to false if your parquet/ipc files already
# contain a spectrum_id column.
add_index_cols: true

@BioGeek BioGeek Jun 8, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would preserve existing spectrum IDs by default. When users load parquet/ipc spectra through the default data_loader=mztab config and the file already contains a spectrum_id, this new default causes _add_index_cols to overwrite those IDs with {file_stem}:{row}. Downstream prediction/FDR exports key results by spectrum_id, so existing stable identifiers are silently lost unless users discover and override this setting. So keep the default false for parquet/ipc or only fill missing IDs.

@JemmaLDaniel JemmaLDaniel Jun 15, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Changed to False and made the docs clearer


residue_remapping: # Used to map Casanovo-specific notations to UNIMOD tokens.
"M+15.995": "M[UNIMOD:35]" # Oxidation
"Q+0.984": "Q[UNIMOD:7]" # Deamidation
Expand Down
Loading
Loading