Merge branch 'releases' into output-IM

.github/workflows/build_and_publish.yml

@@ -12,20 +12,16 @@ jobs:
       - uses: actions/checkout@v4
         with:
           lfs: "true"
-
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
           python-version: "3.11"
-
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
           pip install build
-
       - name: Build sdist
         run: python -m build --sdist --outdir dist
-
       - uses: actions/upload-artifact@v4
         with:
           name: dist-source
@@ -41,18 +37,10 @@ jobs:
       - uses: actions/checkout@v4
         with:
           lfs: "true"
-
-      - uses: actions/setup-python@v5
-        name: Set up Python
-        with:
-          python-version: "3.11"
-
-      - name: Install cibuildwheel
-        run: python -m pip install cibuildwheel>=2
-
       - name: Build wheels
-        run: python -m cibuildwheel --output-dir dist
-
+        uses: pypa/[email protected]
+        with:
+          output-dir: dist
       - uses: actions/upload-artifact@v4
         with:
           name: dist-${{ matrix.os }}

.github/workflows/test.yml

@@ -12,8 +12,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       max-parallel: 4
+      fail-fast: false
       matrix:
-        python-version: ["3.8", "3.9", "3.10", "3.11"]
+        python-version: ["3.9", "3.10", "3.11", "3.12"]
 
     steps:
       - uses: actions/checkout@v4

.readthedocs.yml

@@ -3,7 +3,7 @@ version: 2
 build:
   os: "ubuntu-20.04"
   tools:
-    python: "3.8"
+    python: "3.11"
   jobs:
     post_checkout:
       # Download and uncompress the binary

docs/source/installation.rst

@@ -6,7 +6,7 @@ Pip package
 .. image:: https://flat.badgen.net/badge/install%20with/pip/green
    :target: https://pypi.org/project/ms2pip/
 
-With Python 3.8 or higher, run:
+With Python 3.9 or higher, run:
 
 .. code-block:: bash
 

docs/source/usage.rst

@@ -68,6 +68,14 @@ For instance:
     ms2pip correlate --psm-filetype sage results.sage.tsv spectra.mgf
 
 
+``correlate-single``
+--------------------
+
+Predict spectrum intensities for a single peptide and correlate them with observed intensities from
+an :py:class:`ObservedSpectrum` object. This mode is only available through the Python API, not
+through the command-line interface.
+
+
 ``get-training-data``
 ---------------------
 

ms2pip/core.py

@@ -17,17 +17,17 @@
 from rich.progress import track
 
 import ms2pip.exceptions as exceptions
-from ms2pip import spectrum_output
 from ms2pip._cython_modules import ms2pip_pyx
 from ms2pip._utils.encoder import Encoder
 from ms2pip._utils.feature_names import get_feature_names
+from ms2pip._utils.ion_mobility import IonMobility
 from ms2pip._utils.psm_input import read_psms
 from ms2pip._utils.retention_time import RetentionTime
-from ms2pip._utils.ion_mobility import IonMobility
 from ms2pip._utils.xgb_models import get_predictions_xgb, validate_requested_xgb_model
 from ms2pip.constants import MODELS
 from ms2pip.result import ProcessingResult, calculate_correlations
 from ms2pip.search_space import ProteomeSearchSpace
+from ms2pip.spectrum import ObservedSpectrum
 from ms2pip.spectrum_input import read_spectrum_file
 from ms2pip.spectrum_output import SUPPORTED_FORMATS
 
@@ -291,6 +291,62 @@ def correlate(
     return results
 
 
+def correlate_single(
+    observed_spectrum: ObservedSpectrum,
+    ms2_tolerance: float = 0.02,
+    model: str = "HCD",
+) -> ProcessingResult:
+    """
+    Correlate single observed spectrum with predicted intensities.\f
+
+    Parameters
+    ----------
+    observed_spectrum
+        ObservedSpectrum instance with observed m/z and intensity values and peptidoform.
+    ms2_tolerance
+        MS2 tolerance in Da for observed spectrum peak annotation. By default, 0.02 Da.
+    model
+        Model to use for prediction. Default: "HCD".
+
+    Returns
+    -------
+    result: ProcessingResult
+        Result with theoretical m/z, predicted intensity, observed intensity, and correlation.
+
+    """
+    # Check peptidoform in observed spectrum
+    if not isinstance(observed_spectrum.peptidoform, Peptidoform):
+        raise ValueError("Peptidoform must be set in observed spectrum to correlate.")
+
+    # Annotate spectrum and get target intensities
+    with Encoder.from_peptidoform(observed_spectrum.peptidoform) as encoder:
+        ms2pip_pyx.ms2pip_init(*encoder.encoder_files)
+        enc_peptidoform = encoder.encode_peptidoform(observed_spectrum.peptidoform)
+        targets = ms2pip_pyx.get_targets(
+            enc_peptidoform,
+            observed_spectrum.mz.astype(np.float32),
+            observed_spectrum.intensity.astype(np.float32),
+            float(ms2_tolerance),
+            MODELS[model]["peaks_version"],
+        )
+
+    # Reshape to dict with intensities per ion type
+    ion_types = [it.lower() for it in MODELS[model]["ion_types"]]
+    observed_intensity = {
+        i: np.array(p, dtype=np.float32).clip(min=np.log2(0.001))  # Clip negative intensities
+        for i, p in zip(ion_types, targets)
+    }
+
+    # Predict spectrum and add target intensities
+    result = predict_single(observed_spectrum.peptidoform, model=model)
+    result.observed_intensity = observed_intensity
+
+    # Add correlation
+    calculate_correlations([result])
+
+    return result
+
+
 def get_training_data(
     psms: Union[PSMList, str, Path],
     spectrum_file: Union[str, Path],
@@ -704,19 +760,6 @@ def _add_xgboost_predictions(self, results: List[ProcessingResult]) -> List[Proc
 
         return results
 
-    # TODO IMPLEMENT
-    def write_predictions(
-        self, all_preds: pd.DataFrame, peptides: pd.DataFrame, output_filename: str
-    ):
-        raise NotImplementedError
-        spec_out = spectrum_output.SpectrumOutput(
-            all_preds,
-            peptides,
-            self.params["ms2pip"],
-            output_filename=output_filename,
-        )
-        spec_out.write_results(self.output_formats)
-
 
 def _process_peptidoform(
     psm_index: int,

ms2pip/search_space.py

@@ -160,7 +160,7 @@ class ProteomeSearchSpace(BaseModel):
     min_length: int = 8
     max_length: int = 30
     min_precursor_mz: Optional[float] = 0
-    max_precursor_mz: Optional[float] = np.Inf
+    max_precursor_mz: Optional[float] = np.inf
     cleavage_rule: str = "trypsin"
     missed_cleavages: int = 2
     semi_specific: bool = False
@@ -184,7 +184,7 @@ def __init__(self, **data: Any):
         min_precursor_mz
             Minimum precursor m/z for peptides. Default is 0.
         max_precursor_mz
-            Maximum precursor m/z for peptides. Default is np.Inf.
+            Maximum precursor m/z for peptides. Default is np.inf.
         cleavage_rule
             Cleavage rule for peptide digestion. Default is "trypsin".
         missed_cleavages

pyproject.toml

@@ -31,16 +31,16 @@ classifiers = [
     "Topic :: Scientific/Engineering :: Bio-Informatics",
     "Development Status :: 5 - Production/Stable",
 ]
-requires-python = ">=3.8"
+requires-python = ">=3.9"
 dependencies = [
-    "numpy>=1.16,<2",
-    "pandas>=1,<2",
+    "numpy>=1.25,<3",
+    "pandas>=1,<3",
     "pyarrow",
     "pyteomics>=3.5,<5",
     "tomlkit>=0.5,<1",
     "sqlalchemy>=1.3,<2",
     "click>=7,<9",
-    "xgboost>=1.3,<2",
+    "xgboost>=1.3",
     "lxml>=4",
     "rich>=13",
     "pydantic>=2",
@@ -75,25 +75,23 @@ publication = "https://doi.org/10.1093/nar/gkad335/"
 ms2pip = "ms2pip.__main__:main"
 
 [build-system]
-requires = ["setuptools", "cython", "oldest-supported-numpy"]
+requires = ["setuptools", "cython", "numpy>=2.0"]
 build-backend = "setuptools.build_meta"
 
 [tool.setuptools.packages.find]
 include = ["ms2pip*"]
 
 [tool.black]
 line-length = 99
-target-version = ['py38']
+target-version = ['py39']
 
 [tool.ruff]
 line-length = 99
-target-version = 'py38'
+target-version = 'py39'
 
 [tool.cibuildwheel]
-build = "cp3*-manylinux_x86_64 cp3*-win_amd64 cp3*-macosx_x86_64 cp3*-macosx_arm64"
-skip = "cp36-* cp37-* cp312-*"                                    # EOL / no Numpy wheels available yet for Python 3.12
-manylinux-x86_64-image = "manylinux2014"
-# test-command = "pytest {package}/tests"
+build = "cp3*-manylinux_x86_64 cp3*-musllinux_x86_64 cp3*-win_amd64 cp3*-macosx_x86_64 cp3*-macosx_arm64"
+skip = "cp36-* cp37-* cp38-* cp313-*"                                                                     # No ms2rescore-rs for 3.13
 test-command = "ms2pip --help"
 
 [tool.cibuildwheel.macos]

tests/test_core.py

@@ -1,6 +1,9 @@
+import numpy as np
+from psm_utils import PSM, Peptidoform
 import pandas as pd
 
-from ms2pip.core import get_training_data
+from ms2pip.core import get_training_data, predict_single
+from ms2pip.result import ProcessingResult
 
 
 def _test_get_training_data():
@@ -13,3 +16,33 @@ def _test_get_training_data():
         processes=1
     )
     pd.testing.assert_frame_equal(expected_df, output_df)
+
+def test_predict_single():
+    pep = Peptidoform("ACDE/2")
+    result = predict_single(pep)
+
+    expected = ProcessingResult(
+        psm_index=0,
+        psm=PSM(peptidoform=pep, spectrum_id=0),
+        theoretical_mz={
+            "b": np.array([72.04435, 175.05354, 290.08047], dtype=np.float32),
+            "y": np.array([148.0604, 263.0873, 366.0965], dtype=np.float32),
+        },
+        predicted_intensity={
+            "b": np.array([-9.14031, -7.6102686, -7.746709], dtype=np.float32),
+            "y": np.array([-5.8988147, -5.811797, -7.069088], dtype=np.float32),
+        },
+        observed_intensity=None,
+        correlation=None,
+        feature_vectors=None,
+    )
+
+    assert result.psm_index == expected.psm_index
+    assert result.psm == expected.psm
+    np.testing.assert_array_almost_equal(result.theoretical_mz["b"], expected.theoretical_mz["b"])
+    np.testing.assert_array_almost_equal(result.theoretical_mz["y"], expected.theoretical_mz["y"])
+    np.testing.assert_array_almost_equal(result.predicted_intensity["b"], expected.predicted_intensity["b"])
+    np.testing.assert_array_almost_equal(result.predicted_intensity["y"], expected.predicted_intensity["y"])
+    assert result.observed_intensity == expected.observed_intensity
+    assert result.correlation == expected.correlation
+    assert result.feature_vectors == expected.feature_vectors