feat: Implement CLI

- Add single CLI entrypoint `predict` with option for calibration and fine-tuning

Author: RalfG

deeplc/__main__.py

@@ -1,32 +1,127 @@
 """Main command line interface to DeepLC."""
 
 import logging
-import sys
+from pathlib import Path
 
-LOGGER = logging.getLogger(__name__)
+import click
+import pandas as pd
+from psm_utils.io import READERS, read_file
+from rich.logging import RichHandler
+from rich.traceback import install as install_rich_traceback
 
-# TODO: Add CLI functionality
+import deeplc.core
+from deeplc import __version__
 
+logger = logging.getLogger(__name__)
 
-def _setup_logging(passed_level):
-    log_mapping = {
-        "critical": logging.CRITICAL,
-        "error": logging.ERROR,
-        "warning": logging.WARNING,
-        "info": logging.INFO,
-        "debug": logging.DEBUG,
-    }
+LOGGING_LEVELS = {
+    "DEBUG": logging.DEBUG,
+    "INFO": logging.INFO,
+    "WARNING": logging.WARNING,
+    "ERROR": logging.ERROR,
+    "CRITICAL": logging.CRITICAL,
+}
 
-    if passed_level.lower() not in log_mapping:
-        print(
-            "Invalid log level. Should be one of the following: ",
-            ", ".join(log_mapping.keys()),
-        )
-        exit(1)
+PSM_FILETYPES = list(READERS.keys())
 
+
+def _infer_output_name(input_filename: str, output_name: str | None = None) -> Path:
+    """Infer output filename from input filename if not provided."""
+    if output_name:
+        return Path(output_name)
+    else:
+        input_path = Path(input_filename)
+        return input_path.with_name(input_path.stem + "_deeplc_predictions").with_suffix(".csv")
+
+
+def _read_psm_file(psms: str, psm_filetype: str | None = None):
+    """Read a PSM file and return a PSMList."""
+    logger.info(f"Reading PSM file: {psms}")
+    kwargs = {"filetype": psm_filetype} if psm_filetype else {}
+    return read_file(psms, **kwargs)
+
+
+def _write_predictions(psm_list, predictions, output_path: Path):
+    """Write predictions to a CSV file."""
+    df = pd.DataFrame(
+        {
+            "peptidoform": [str(psm.peptidoform) for psm in psm_list],
+            "observed_rt": [psm.retention_time for psm in psm_list],
+            "predicted_rt": predictions,
+        }
+    )
+    logger.info(f"Writing predictions to {output_path}")
+    df.to_csv(output_path, index=False)
+
+
+@click.group()
+@click.option(
+    "--logging-level",
+    "-l",
+    type=click.Choice(LOGGING_LEVELS.keys()),
+    default="INFO",
+    help="Set the logging level.",
+)
+@click.version_option(version=__version__)
+def cli(logging_level, **kwargs):
+    """DeepLC: Retention time prediction for peptides carrying any modification."""
+    install_rich_traceback(show_locals=True)
     logging.basicConfig(
-        stream=sys.stdout,
-        format="%(asctime)s - %(levelname)s - %(message)s",
+        format="%(message)s",
         datefmt="%Y-%m-%d %H:%M:%S",
-        level=log_mapping[passed_level.lower()],
+        level=LOGGING_LEVELS[logging_level],
+        handlers=[RichHandler(rich_tracebacks=True, show_level=True, show_path=False)],
     )
+
+
+@cli.command()
+@click.argument("psms", required=True, type=click.Path(exists=True, dir_okay=False))
+@click.option("--psm-filetype", "-t", type=click.Choice(PSM_FILETYPES), default=None)
+@click.option(
+    "--reference", type=click.Path(exists=True, dir_okay=False), default=None,
+    help="Reference PSM file for calibration or fine-tuning.",
+)
+@click.option(
+    "--reference-filetype", "-r", type=click.Choice(PSM_FILETYPES), default=None,
+    help="File type for the reference file. Inferred if not provided.",
+)
+@click.option(
+    "--finetune", is_flag=True, default=False,
+    help="Fine-tune the model to the reference before predicting. Requires --reference.",
+)
+@click.option("--output", "-o", type=str, default=None, help="Output file path.")
+@click.option("--model", "-m", type=click.Path(exists=True, dir_okay=False), default=None)
+def predict(psms, psm_filetype, reference, reference_filetype, finetune, output, model):
+    """Predict retention times, optionally calibrating or fine-tuning to a reference."""
+    if finetune and not reference:
+        raise click.UsageError("--finetune requires --reference.")
+
+    psm_list = _read_psm_file(psms, psm_filetype)
+    output_path = _infer_output_name(psms, output)
+
+    if reference:
+        psm_list_reference = _read_psm_file(reference, reference_filetype)
+        if finetune:
+            predictions = deeplc.core.finetune_and_predict(
+                psm_list=psm_list,
+                psm_list_reference=psm_list_reference,
+                model=model,
+            )
+        else:
+            predictions = deeplc.core.predict_and_calibrate(
+                psm_list=psm_list,
+                psm_list_reference=psm_list_reference,
+                model=model,
+            )
+    else:
+        predictions = deeplc.core.predict(psm_list=psm_list, model=model)
+
+    _write_predictions(psm_list, predictions, output_path)
+
+
+def main():
+    cli()
+
+
+if __name__ == "__main__":
+    main()

pyproject.toml

@@ -40,11 +40,12 @@ dependencies = [
     "scikit-learn>=1.2.0,<2",
     "psm_utils>=1.5,<2",
     "onnx2torch>=1.5,<2",
+    "click>=8,<9",
+    "rich>=13,<15",
 ]
 
-# [project.scripts]
-# deeplc = "deeplc.__main__:main"
-# deeplc-gui = "deeplc_gui.__main__:main"
+[project.scripts]
+deeplc = "deeplc.__main__:main"
 
 [project.urls]
 GitHub = "https://github.com/compomics/deeplc"