ms2pip - pypi Package Compare versions

Comparing version 4.1.0 to 4.1.1

ms2pip.egg-info/PKG-INFO

@@ -1,7 +0,7 @@
-Metadata-Version: 2.2
+Metadata-Version: 2.4
 Name: ms2pip
-Version: 4.1.0
+Version: 4.1.1
 Summary: MS2PIP: Accurate and versatile peptide fragmentation spectrum prediction.
-Author: Ana Sílvia C. Silva
-Author-email: Ralf Gabriels <[e-mail]>, Sven Degroeve <[e-mail]>, Arthur Declercq <[e-mail]>, Kevin Velghe <[e-mail]>
+Author: Sven Degroeve, Ana Sílvia C. Silva
+Author-email: Ralf Gabriels <[e-mail]>, Arthur Declercq <[e-mail]>, Kevin Velghe <[e-mail]>
 License:                                  Apache License
@@ -230,4 +230,3 @@ Version 2.0, January 2004
 Requires-Dist: pyteomics<5,>=3.5
-Requires-Dist: tomlkit<1,>=0.5
-Requires-Dist: sqlalchemy<2,>=1.3
+Requires-Dist: sqlalchemy<3,>=1.4
 Requires-Dist: click<9,>=7
@@ -258,2 +257,3 @@ Requires-Dist: xgboost<3,>=1.3
 Requires-Dist: sphinx-click; extra == "docs"
+Dynamic: license-file
 
@@ -260,0 +260,0 @@ .. image:: https://github.com/compomics/ms2pip_c/raw/releases/img/ms2pip_logo_1000px.png

ms2pip.egg-info/requires.txt

@@ -5,4 +5,3 @@ numpy<3,>=1.25
 pyteomics<5,>=3.5
-tomlkit<1,>=0.5
-sqlalchemy<2,>=1.3
+sqlalchemy<3,>=1.4
 click<9,>=7
@@ -9,0 +8,0 @@ xgboost<3,>=1.3

ms2pip/__init__.py

 # isort: skip_file
 """MS2PIP: Accurate and versatile peptide fragmentation spectrum prediction."""
 
-__version__ = "4.1.0"
+__version__ = "4.1.1"
 
@@ -6,0 +6,0 @@ from warnings import filterwarnings

ms2pip/_utils/dlib.py

@@ -21,2 +21,3 @@ """Database configuration for EncyclopeDIA DLIB SQLite format."""
 from sqlalchemy.dialects.sqlite import BLOB
+from sqlalchemy.engine import Connection
 
@@ -27,7 +28,8 @@ DLIB_VERSION = "0.1.14"
 class CompressedArray(TypeDecorator):
-    """ Sqlite-like does not support arrays.
-        Let's use a custom type decorator.
+    """Sqlite-like does not support arrays.
+    Let's use a custom type decorator.
 
-        See http://docs.sqlalchemy.org/en/latest/core/types.html#sqlalchemy.types.TypeDecorator
+    See http://docs.sqlalchemy.org/en/latest/core/types.html#sqlalchemy.types.TypeDecorator
     """
+
     impl = BLOB
@@ -54,49 +56,53 @@
 
-big_float = numpy.dtype('>f4')
-big_double = numpy.dtype('>f8')
+big_float = numpy.dtype(">f4")
+big_double = numpy.dtype(">f8")
 
 Entry = Table(
-    'entries',
+    "entries",
     metadata,
-    Column('PrecursorMz', Float, nullable=False, index=True),
-    Column('PrecursorCharge', Integer, nullable=False),
-    Column('PeptideModSeq', String, nullable=False),
-    Column('PeptideSeq', String, nullable=False, index=True),
-    Column('Copies', Integer, nullable=False),
-    Column('RTInSeconds', Float, nullable=False),
-    Column('Score', Float, nullable=False),
-    Column('MassEncodedLength', Integer, nullable=False),
-    Column('MassArray', CompressedArray(big_double), nullable=False),
-    Column('IntensityEncodedLength', Integer, nullable=False),
-    Column('IntensityArray', CompressedArray(big_float), nullable=False),
-    Column('CorrelationEncodedLength', Integer, nullable=True),
-    Column('CorrelationArray', CompressedArray(big_float), nullable=True),
-    Column('RTInSecondsStart', Float, nullable=True),
-    Column('RTInSecondsStop', Float, nullable=True),
-    Column('MedianChromatogramEncodedLength', Integer, nullable=True),
-    Column('MedianChromatogramArray', CompressedArray(big_float), nullable=True),
-    Column('SourceFile', String, nullable=False),
+    Column("PrecursorMz", Float, nullable=False, index=True),
+    Column("PrecursorCharge", Integer, nullable=False),
+    Column("PeptideModSeq", String, nullable=False),
+    Column("PeptideSeq", String, nullable=False, index=True),
+    Column("Copies", Integer, nullable=False),
+    Column("RTInSeconds", Float, nullable=False),
+    Column("Score", Float, nullable=False),
+    Column("MassEncodedLength", Integer, nullable=False),
+    Column("MassArray", CompressedArray(big_double), nullable=False),
+    Column("IntensityEncodedLength", Integer, nullable=False),
+    Column("IntensityArray", CompressedArray(big_float), nullable=False),
+    Column("CorrelationEncodedLength", Integer, nullable=True),
+    Column("CorrelationArray", CompressedArray(big_float), nullable=True),
+    Column("RTInSecondsStart", Float, nullable=True),
+    Column("RTInSecondsStop", Float, nullable=True),
+    Column("MedianChromatogramEncodedLength", Integer, nullable=True),
+    Column("MedianChromatogramArray", CompressedArray(big_float), nullable=True),
+    Column("SourceFile", String, nullable=False),
 )
 
-Index('ix_entries_PeptideModSeq_PrecursorCharge_SourceFile', Entry.c.PeptideModSeq, Entry.c.PrecursorCharge, Entry.c.SourceFile)
+Index(
+    "ix_entries_PeptideModSeq_PrecursorCharge_SourceFile",
+    Entry.c.PeptideModSeq,
+    Entry.c.PrecursorCharge,
+    Entry.c.SourceFile,
+)
 
 PeptideToProtein = Table(
-    'peptidetoprotein',
+    "peptidetoprotein",
     metadata,
-    Column('PeptideSeq', String, nullable=False, index=True),
-    Column('isDecoy', Boolean, nullable=True),
-    Column('ProteinAccession', String, nullable=False, index=True),
+    Column("PeptideSeq", String, nullable=False, index=True),
+    Column("isDecoy", Boolean, nullable=True),
+    Column("ProteinAccession", String, nullable=False, index=True),
 )
 
 Metadata = Table(
-    'metadata',
+    "metadata",
     metadata,
-    Column('Key', String, nullable=False, index=True),
-    Column('Value', String, nullable=False),
+    Column("Key", String, nullable=False, index=True),
+    Column("Value", String, nullable=False),
 )
 
 
-def open_sqlite(filename: Union[str, Path]) -> sqlalchemy.engine.Connection:
+def open_sqlite(filename: Union[str, Path]) -> Connection:
     engine = sqlalchemy.create_engine(f"sqlite:///{filename}")
-    metadata.bind = engine
     return engine.connect()

ms2pip/search_space.py

@@ -18,3 +18,3 @@ """
      "min_length": 8,
-     "max_length": 3,
+     "max_length": 30,
      "cleavage_rule": "trypsin",
@@ -21,0 +21,0 @@ "missed_cleavages": 2,

ms2pip/spectrum_output.py

@@ -50,2 +50,3 @@ """
 from pathlib import Path
+from os import PathLike
 from time import localtime, strftime
@@ -57,3 +58,4 @@ from typing import Any, Dict, Generator, List, Optional, Union
 from pyteomics import proforma
-from sqlalchemy import engine, select
+from sqlalchemy import select
+from sqlalchemy.engine import Connection
 
@@ -67,3 +69,3 @@ from ms2pip._utils import dlib
 def write_spectra(
-    filename: Union[str, Path],
+    filename: Union[str, PathLike],
     processing_results: List[ProcessingResult],
@@ -98,3 +100,3 @@ file_format: str = "tsv",
 
-    def __init__(self, filename: Union[str, Path], write_mode: str = "w"):
+    def __init__(self, filename: Union[str, PathLike], write_mode: str = "w"):
         self.filename = Path(filename).with_suffix(self.suffix)
@@ -473,3 +475,3 @@ self.write_mode = write_mode
 
-    def __init__(self, filename: Union[str, Path], write_mode: str = "w"):
+    def __init__(self, filename: Union[str, PathLike], write_mode: str = "w"):
         super().__init__(filename, write_mode)
@@ -626,3 +628,3 @@ self.ssl_file = self.filename.with_suffix(self.ssl_suffix)
     @staticmethod
-    def _get_last_ssl_scan_number(ssl_file: Union[str, Path, StringIO]):
+    def _get_last_ssl_scan_number(ssl_file: Union[str, PathLike, StringIO]):
         """Read scan number of last line in a Bibliospec SSL file."""
@@ -662,3 +664,3 @@ if isinstance(ssl_file, StringIO):
         connection = self._file_object
-        dlib.metadata.create_all()
+        dlib.metadata.create_all(connection.engine)
         self._write_metadata(connection)
@@ -692,7 +694,7 @@ self._write_entries(processing_results, connection, self.filename)
     @staticmethod
-    def _write_metadata(connection: engine.Connection):
+    def _write_metadata(connection: Connection):
         """Write metadata to DLIB SQLite file."""
         with connection.begin():
             version = connection.execute(
-                select([dlib.Metadata.c.Value]).where(dlib.Metadata.c.Key == "version")
+                select(dlib.Metadata.c.Value).where(dlib.Metadata.c.Key == "version")
             ).scalar()
@@ -710,4 +712,4 @@ if version is None:
         processing_results: List[ProcessingResult],
-        connection: engine.Connection,
-        output_filename: str,
+        connection: Connection,
+        output_filename: Union[str, PathLike],
     ):
@@ -742,3 +744,3 @@ """Write spectra to DLIB SQLite file."""
     @staticmethod
-    def _write_peptide_to_protein(results: List[ProcessingResult], connection: engine.Connection):
+    def _write_peptide_to_protein(results: List[ProcessingResult], connection: Connection):
         """Write peptide-to-protein mappings to DLIB SQLite file."""
@@ -756,3 +758,3 @@ peptide_to_proteins = {
             for peptide_to_protein in connection.execute(
-                dlib.PeptideToProtein.select().where(
+                select(dlib.PeptideToProtein).where(
                     dlib.PeptideToProtein.c.ProteinAccession.in_(proteins)
@@ -759,0 +761,0 @@ )

PKG-INFO

@@ -1,7 +0,7 @@
-Metadata-Version: 2.2
+Metadata-Version: 2.4
 Name: ms2pip
-Version: 4.1.0
+Version: 4.1.1
 Summary: MS2PIP: Accurate and versatile peptide fragmentation spectrum prediction.
-Author: Ana Sílvia C. Silva
-Author-email: Ralf Gabriels <[e-mail]>, Sven Degroeve <[e-mail]>, Arthur Declercq <[e-mail]>, Kevin Velghe <[e-mail]>
+Author: Sven Degroeve, Ana Sílvia C. Silva
+Author-email: Ralf Gabriels <[e-mail]>, Arthur Declercq <[e-mail]>, Kevin Velghe <[e-mail]>
 License:                                  Apache License
@@ -230,4 +230,3 @@ Version 2.0, January 2004
 Requires-Dist: pyteomics<5,>=3.5
-Requires-Dist: tomlkit<1,>=0.5
-Requires-Dist: sqlalchemy<2,>=1.3
+Requires-Dist: sqlalchemy<3,>=1.4
 Requires-Dist: click<9,>=7
@@ -258,2 +257,3 @@ Requires-Dist: xgboost<3,>=1.3
 Requires-Dist: sphinx-click; extra == "docs"
+Dynamic: license-file
 
@@ -260,0 +260,0 @@ .. image:: https://github.com/compomics/ms2pip_c/raw/releases/img/ms2pip_logo_1000px.png

pyproject.toml

@@ -19,3 +19,3 @@ [project]
     { name = "Ralf Gabriels", email = "[e-mail]" },
-    { name = "Sven Degroeve", email = "[e-mail]" },
+    { name = "Sven Degroeve" },
     { name = "Arthur Declercq", email = "[e-mail]" },
@@ -41,4 +41,3 @@ { name = "Kevin Velghe", email = "[e-mail]" },
     "pyteomics>=3.5,<5",
-    "tomlkit>=0.5,<1",
-    "sqlalchemy>=1.3,<2",
+    "sqlalchemy>=1.4,<3",
     "click>=7,<9",
@@ -95,7 +94,10 @@ "xgboost>=1.3,<3",
 [tool.cibuildwheel]
-build = "cp39*-manylinux_x86_64 cp39*-win_amd64 cp39*-macosx_x86_64 cp39*-macosx_arm64"
+build = "cp3*-manylinux_x86_64 cp3*-win_amd64 cp3*-macosx_x86_64 cp3*-macosx_arm64"
 skip = "cp36-* cp37-* cp38-*"
 test-command = "ms2pip --help"
 
+# Prevent building from source for packages with complex C/C++ dependencies
+environment = { PIP_ONLY_BINARY = "pyarrow,pandas,numpy,lxml,xgboost" }
+
 [tool.cibuildwheel.macos]
 before-all = "brew install libomp"

tests/test_spectrum_output.py

@@ -1,6 +0,12 @@
-from psm_utils import Peptidoform
+import tempfile
+from pathlib import Path
 
-from ms2pip.spectrum_output import MSP, Bibliospec, DLIB
+import numpy as np
+import pytest
+from psm_utils import PSM, Peptidoform
 
+from ms2pip.result import ProcessingResult
+from ms2pip.spectrum_output import DLIB, MSP, Bibliospec
 
+
 class TestMSP:
@@ -48,1 +54,224 @@ def test__format_modification_string(self):
             assert DLIB._format_modified_sequence(Peptidoform(test_in)) == expected_out
+
+    def test_dlib_database_creation(self):
+        """Test that DLIB file creation works with SQLAlchemy (integration test)."""
+        # Create test data
+        pep = Peptidoform("ACDE/2")
+        psm = PSM(
+            peptidoform=pep,
+            spectrum_id=1,
+            retention_time=100.0,
+            protein_list=["PROT1", "PROT2"],
+        )
+        result = ProcessingResult(
+            psm_index=0,
+            psm=psm,
+            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([0.1, 0.5, 0.3], dtype=np.float32),
+                "y": np.array([0.8, 0.6, 0.2], dtype=np.float32),
+            },
+            observed_intensity=None,
+            correlation=None,
+            feature_vectors=None,
+        )
+
+        # Write DLIB file
+        with tempfile.TemporaryDirectory() as tmpdir:
+            dlib_file = Path(tmpdir) / "test.dlib"
+            with DLIB(dlib_file) as writer:
+                writer.write([result])
+
+            # Verify file was created
+            assert dlib_file.exists()
+
+            # Verify database structure and content using SQLAlchemy
+            from ms2pip._utils import dlib as dlib_module
+
+            connection = dlib_module.open_sqlite(dlib_file)
+            try:
+                # Test that metadata table exists and has version
+                from sqlalchemy import select
+
+                version = connection.execute(
+                    select(dlib_module.Metadata.c.Value).where(
+                        dlib_module.Metadata.c.Key == "version"
+                    )
+                ).scalar()
+                assert version == dlib_module.DLIB_VERSION
+
+                # Test that Entry table has data (select specific columns to avoid nullable CompressedArray)
+                from sqlalchemy import func
+
+                entry_count = connection.execute(
+                    select(func.count()).select_from(dlib_module.Entry)
+                ).scalar()
+                assert entry_count == 1
+
+                # Select specific non-nullable columns
+                entry = connection.execute(
+                    select(
+                        dlib_module.Entry.c.PeptideSeq,
+                        dlib_module.Entry.c.PrecursorCharge,
+                        dlib_module.Entry.c.RTInSeconds,
+                        dlib_module.Entry.c.MassArray,
+                        dlib_module.Entry.c.IntensityArray,
+                    )
+                ).fetchone()
+                assert entry.PeptideSeq == "ACDE"
+                assert entry.PrecursorCharge == 2
+                assert entry.RTInSeconds == 100.0
+                assert len(entry.MassArray) == 6  # 3 b-ions + 3 y-ions
+                assert len(entry.IntensityArray) == 6
+
+                # Test that PeptideToProtein table has data
+                peptide_to_proteins = connection.execute(
+                    select(dlib_module.PeptideToProtein)
+                ).fetchall()
+                assert len(peptide_to_proteins) == 2
+                proteins = {p.ProteinAccession for p in peptide_to_proteins}
+                assert proteins == {"PROT1", "PROT2"}
+                assert all(p.PeptideSeq == "ACDE" for p in peptide_to_proteins)
+            finally:
+                connection.close()
+
+    def test_dlib_multiple_results(self):
+        """Test writing multiple ProcessingResults to DLIB file."""
+        # Create multiple test results
+        results = []
+        for i, seq in enumerate(["ACDE/2", "PEPTIDE/2", "TESTK/2"]):
+            pep = Peptidoform(seq)
+            psm = PSM(
+                peptidoform=pep,
+                spectrum_id=i,
+                retention_time=100.0 + i * 10,
+                protein_list=[f"PROT{i}"],
+            )
+            result = ProcessingResult(
+                psm_index=i,
+                psm=psm,
+                theoretical_mz={
+                    "b": np.array([72.04435, 175.05354], dtype=np.float32),
+                    "y": np.array([148.0604, 263.0873], dtype=np.float32),
+                },
+                predicted_intensity={
+                    "b": np.array([0.1, 0.5], dtype=np.float32),
+                    "y": np.array([0.8, 0.6], dtype=np.float32),
+                },
+                observed_intensity=None,
+                correlation=None,
+                feature_vectors=None,
+            )
+            results.append(result)
+
+        # Write DLIB file
+        with tempfile.TemporaryDirectory() as tmpdir:
+            dlib_file = Path(tmpdir) / "test_multiple.dlib"
+            with DLIB(dlib_file) as writer:
+                writer.write(results)
+
+            # Verify all entries were written
+            from sqlalchemy import select
+
+            from ms2pip._utils import dlib as dlib_module
+
+            connection = dlib_module.open_sqlite(dlib_file)
+            try:
+                # Select specific columns to avoid nullable CompressedArray
+                entries = connection.execute(
+                    select(
+                        dlib_module.Entry.c.PeptideSeq,
+                        dlib_module.Entry.c.RTInSeconds,
+                    )
+                ).fetchall()
+                assert len(entries) == 3
+
+                peptides = {e.PeptideSeq for e in entries}
+                assert peptides == {"ACDE", "PEPTIDE", "TESTK"}
+
+                # Verify retention times
+                rt_values = {e.RTInSeconds for e in entries}
+                assert rt_values == {100.0, 110.0, 120.0}
+            finally:
+                connection.close()
+
+    def test_dlib_sqlalchemy_select_syntax(self):
+        """Test that SQLAlchemy v2 select() syntax works correctly."""
+        # This test specifically verifies the SQLAlchemy v2 compatibility changes
+        pep = Peptidoform("ACDE/2")
+        psm = PSM(
+            peptidoform=pep,
+            spectrum_id=1,
+            retention_time=100.0,
+            protein_list=["PROT1"],
+        )
+        result = ProcessingResult(
+            psm_index=0,
+            psm=psm,
+            theoretical_mz={
+                "b": np.array([72.04435], dtype=np.float32),
+                "y": np.array([148.0604], dtype=np.float32),
+            },
+            predicted_intensity={
+                "b": np.array([0.5], dtype=np.float32),
+                "y": np.array([0.8], dtype=np.float32),
+            },
+            observed_intensity=None,
+            correlation=None,
+            feature_vectors=None,
+        )
+
+        with tempfile.TemporaryDirectory() as tmpdir:
+            dlib_file = Path(tmpdir) / "test_sqlalchemy.dlib"
+            with DLIB(dlib_file) as writer:
+                writer.write([result])
+
+            # Test the specific SQLAlchemy operations that were modified
+            from sqlalchemy import select
+
+            from ms2pip._utils import dlib as dlib_module
+
+            connection = dlib_module.open_sqlite(dlib_file)
+            try:
+                # Test select(Table) syntax (changed from Table.select())
+                peptide_to_protein_results = connection.execute(
+                    select(dlib_module.PeptideToProtein).where(
+                        dlib_module.PeptideToProtein.c.ProteinAccession == "PROT1"
+                    )
+                ).fetchall()
+                assert len(peptide_to_protein_results) == 1
+                assert peptide_to_protein_results[0].PeptideSeq == "ACDE"
+
+                # Test select(column) syntax (changed from select([column]))
+                version = connection.execute(
+                    select(dlib_module.Metadata.c.Value).where(
+                        dlib_module.Metadata.c.Key == "version"
+                    )
+                ).scalar()
+                assert version is not None
+                assert version == dlib_module.DLIB_VERSION
+            finally:
+                connection.close()
+
+    def test_dlib_missing_retention_time(self):
+        """Test that DLIB writing raises error when retention time is missing."""
+        pep = Peptidoform("ACDE/2")
+        psm = PSM(peptidoform=pep, spectrum_id=1)  # No retention_time
+        result = ProcessingResult(
+            psm_index=0,
+            psm=psm,
+            theoretical_mz={"b": np.array([72.04435], dtype=np.float32)},
+            predicted_intensity={"b": np.array([0.5], dtype=np.float32)},
+            observed_intensity=None,
+            correlation=None,
+            feature_vectors=None,
+        )
+
+        with tempfile.TemporaryDirectory() as tmpdir:
+            dlib_file = Path(tmpdir) / "test_no_rt.dlib"
+            with pytest.raises(ValueError, match="Retention time required"):
+                with DLIB(dlib_file) as writer:
+                    writer.write([result])

Alert delta unavailable

Currently unable to show alert delta for PyPI packages.

Improved metrics

Total package byte prevSize

35515142

0.03%

Lines of code

4224

5.23%