Release/0.3.2

.cookietemple.yml

@@ -15,5 +15,5 @@ full_name: Mario Picciani
 email: [email protected]
 project_name: spectrum_io
 project_short_description: IO related functionalities for oktoberfest.
-version: 0.3.1
+version: 0.3.2
 license: MIT

.github/release-drafter.yml

@@ -1,5 +1,5 @@
-name-template: "0.3.1 🌈" # <<COOKIETEMPLE_FORCE_BUMP>>
-tag-template: 0.3.1 # <<COOKIETEMPLE_FORCE_BUMP>>
+name-template: "0.3.2 🌈" # <<COOKIETEMPLE_FORCE_BUMP>>
+tag-template: 0.3.2 # <<COOKIETEMPLE_FORCE_BUMP>>
 exclude-labels:
     - "skip-changelog"
 

cookietemple.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.1
+current_version = 0.3.2
 
 [bumpversion_files_whitelisted]
 init_file = spectrum_io/__init__.py

docs/conf.py

@@ -53,9 +53,9 @@
 # the built documents.
 #
 # The short X.Y version.
-version = "0.3.1"
+version = "0.3.2"
 # The full version, including alpha/beta/rc tags.
-release = "0.3.1"
+release = "0.3.2"
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "spectrum_io"
-version = "0.3.1"  # <<COOKIETEMPLE_FORCE_BUMP>>
+version = "0.3.2"  # <<COOKIETEMPLE_FORCE_BUMP>>
 description = "IO related functionalities for oktoberfest."
 authors = ["Mario Picciani <[email protected]>"]
 license = "MIT"
@@ -30,7 +30,7 @@ pymzml = "^2.5.0"
 pyteomics = "^4.3.3"
 lxml= '^4.5.2'
 tables = "^3.6.1"
-spectrum-fundamentals = "^0.4.0"
+spectrum-fundamentals = ">=0.4.3,<0.5.0"
 
 [tool.poetry.dev-dependencies]
 pytest = ">=6.2.3"
@@ -99,3 +99,5 @@ requires = [
     "wheel"
 ]
 build-backend = "poetry.core.masonry.api"
+
+[tool.setuptools_scm]

spectrum_io/__init__.py

@@ -2,7 +2,7 @@
 
 __author__ = "Mario Picciani"
 __email__ = "[email protected]"
-__version__ = "0.3.1"
+__version__ = "0.3.2"
 
 import logging
 import logging.handlers

spectrum_io/__main__.py

@@ -5,7 +5,7 @@
 
 
 @click.command()
-@click.version_option(version="0.3.1", message=click.style("spectrum_io Version: 0.3.1"))
+@click.version_option(version="0.3.2", message=click.style("spectrum_io Version: 0.3.2"))
 def main() -> None:
     """spectrum_io."""
 

spectrum_io/search_result/maxquant.py

@@ -82,7 +82,7 @@ def update_columns_for_prosit(df: pd.DataFrame, tmt_labeled: str) -> pd.DataFram
             logger.info("Adding TMT fixed modifications")
             df["MODIFIED_SEQUENCE"] = maxquant_to_internal(
                 df["MODIFIED_SEQUENCE"].to_numpy(),
-                fixed_mods={"C": "C[UNIMOD:4]", "^_": f"_{unimod_tag}", "K": f"K{unimod_tag}"},
+                fixed_mods={"C": "C[UNIMOD:4]", "^_": f"_{unimod_tag}-", "K": f"K{unimod_tag}"},
             )
             df["MASS"] = df.apply(lambda x: MaxQuant.add_tmt_mod(x.MASS, x.MODIFIED_SEQUENCE, unimod_tag), axis=1)
             if "msa" in tmt_labeled:

spectrum_io/spectral_library/__init__.py

@@ -4,6 +4,7 @@
 from . import digest
 from .dlib import DLib
 from .msp import MSP
+from .spectral_library import SpectralLibrary
 from .spectronaut import Spectronaut
 
 logger = logging.getLogger(__name__)

spectrum_io/spectral_library/dlib.py

@@ -5,6 +5,7 @@
 
 import numpy as np
 import pandas as pd
+from spectrum_fundamentals.constants import PARTICLE_MASSES
 from spectrum_fundamentals.mod_string import internal_to_mod_mass, internal_without_mods
 
 from .spectral_library import SpectralLibrary
@@ -25,62 +26,9 @@
 class DLib(SpectralLibrary):
     """Main to init a DLib obj."""
 
-    def __init__(
-        self,
-        precursor_mz: Union[List[float], np.ndarray],
-        precursor_charges: Union[List[int], np.ndarray],
-        modified_sequences: List[str],
-        retention_times: Union[List[float], np.ndarray],
-        fragmentmz: List[np.ndarray],
-        intensities: List[np.ndarray],
-        path: Union[str, Path],
-        min_intensity_threshold: Optional[float] = 0.05,
-    ):
+    def _calculate_masked_values(self, fragmentmz: List[np.ndarray], intensities: List[np.ndarray]):
         """
-        Initializer for the DLib class.
-
-        :param precursor_mz: precursor mass to charge ratios
-        :param precursor_charges: precurosr charges
-        :param modified_sequences: modified sequences in internal format
-        :param retention_times: retention times
-        :param fragmentmz: mass to charge ratio of fragments
-        :param intensities: intensities
-        :param path: path to the file the dlib is written to
-        :param min_intensity_threshold: minimal intensity required when masking fragmentmz and intensities
-        """
-        self.path = path
-        self.create_database(self.path)
-
-        # gather all values for the entries table and create pandas DataFrame
-        masked_values = self._calculate_masked_values(fragmentmz, intensities, min_intensity_threshold)
-        mass_mod_sequences = internal_to_mod_mass(modified_sequences)
-        sequences = internal_without_mods(modified_sequences)
-        data_list = [*masked_values, precursor_charges, mass_mod_sequences, sequences, retention_times, precursor_mz]
-        self.entries = pd.DataFrame(dict(zip(DLIB_COL_NAMES, data_list)))
-
-        # hardcoded entries that we currently not use.
-        # Visit https://bitbucket.org/searleb/encyclopedia/wiki/EncyclopeDIA%20File%20Formats for dlib specs
-        self.entries["Copies"] = 1  # this is hardcorded for now and unused
-        self.entries["Score"] = 0
-        self.entries["CorrelationEncodedLength"] = None
-        self.entries["CorrelationArray"] = None
-        self.entries["RTInSecondsStart"] = None
-        self.entries["RTInSecondsStop"] = None
-        self.entries["MedianChromatogramEncodedLength"] = None
-        self.entries["MedianChromatogramArray"] = None
-        self.entries["SourceFile"] = "Prosit"
-
-        # gather all values for the p2p table and create pandas DataFrame
-        self.p2p = pd.DataFrame({"PeptideSeq": sequences, "isDecoy": False, "ProteinAccession": "unknown"})
-
-    @staticmethod
-    def _calculate_masked_values(
-        fragmentmz: List[np.ndarray],
-        intensities: List[np.ndarray],
-        intensity_min_threshold: Optional[float] = 0.05,
-    ):
-        """
-        Internal function called during __init__ that masks, filters, byte encodes, swaps and compresses fragmentmz \
+        Internal function that masks, filters, byte encodes, swaps and compresses fragmentmz \
         and intensities.
 
         This will produce the data for the following columns in this order:
@@ -90,7 +38,6 @@
             - 'IntensityEncodedLength'.
         :param fragmentmz: fragmentmz provided in __init__
         :param intensities: intensities provided in __init__
-        :param intensity_min_threshold: minimum threshold for tge intensity; default=0.05
         :return: 4 lists as described above
         """
         mz_bytes_list = []
@@ -99,8 +46,7 @@
         i_lengths = []
         for mz, i in zip(fragmentmz, intensities):
             # mask to only existing peaks
-            mask = i >= intensity_min_threshold
-            print(mask)
+            mask = i >= self.min_intensity_threshold
             sort_index = np.argsort(mz[mask])
             masked_mz_ordered = mz[mask][sort_index]
             masked_i_ordered = i[mask][sort_index]
@@ -165,14 +111,10 @@
         c.execute(sql_insert_meta, ["staleProteinMapping", "true"])
         conn.commit()
 
-    def write(self, chunksize: Optional[Union[None, int]]):
-        """
-        Writes the entries ad p2p table to file.
-
-        :param chunksize: optional size of chunks to insert at once
-        """
-        self._write_entries(index=False, if_exists="append", method="multi", chunksize=chunksize)
-        self._write_p2p(index=False, if_exists="append", method="multi", chunksize=chunksize)
+    def write(self):
+        """Writes the entries ad p2p table to file."""
+        self._write_entries(index=False, if_exists="append", method="multi", chunksize=self.chunksize)
+        self._write_p2p(index=False, if_exists="append", method="multi", chunksize=self.chunksize)
 
     def _write_entries(self, *args, **kwargs):
         """
@@ -181,7 +123,7 @@
         :param args: forwarded to pandas.to_sql
         :param kwargs: forwarded to pandas.to_sql
         """
-        conn = sqlite3.connect(self.path)
+        conn = sqlite3.connect(self.out_path)
         self.entries.to_sql(name="entries", con=conn, *args, **kwargs)
         conn.commit()
 
@@ -192,10 +134,50 @@
         :param args: forwarded to pandas.to_sql
         :param kwargs: forwarded to pandas.to_sql
         """
-        conn = sqlite3.connect(self.path)
+        conn = sqlite3.connect(self.out_path)
         self.p2p.to_sql(name="peptidetoprotein", con=conn, *args, **kwargs)
         conn.commit()
 
     def prepare_spectrum(self):
-        """Prepare spectrum."""
-        pass
+        """Converts grpc output and metadata dataframe into dlib format."""
+        # precursor_mz: Union[List[float], np.ndarray],
+        # precursor_charges: Union[List[int], np.ndarray],
+        # modified_sequences: List[str],
+        # retention_times: Union[List[float], np.ndarray],
+        # fragmentmz: List[np.ndarray],
+        # intensities: List[np.ndarray],
+
+        intensities = self.grpc_output[list(self.grpc_output)[0]]["intensity"]
+        fragment_mz = self.grpc_output[list(self.grpc_output)[0]]["fragmentmz"]
+        # annotation = self.grpc_output[list(self.grpc_output)[0]]["annotation"]
+        irt = self.grpc_output[list(self.grpc_output)[1]]
+        retention_times = irt.flatten()
+        modified_sequences = self.spectra_input["MODIFIED_SEQUENCE"]
+
+        precursor_charges = self.spectra_input["PRECURSOR_CHARGE"]
+        precursor_masses = self.spectra_input["MASS"]
+        precursor_mz = (precursor_masses + (precursor_charges * PARTICLE_MASSES["PROTON"])) / precursor_charges
+
+        self.create_database(self.out_path)
+
+        # gather all values for the entries table and create pandas DataFrame
+        masked_values = self._calculate_masked_values(fragment_mz, intensities)
+        mass_mod_sequences = internal_to_mod_mass(modified_sequences)
+        sequences = internal_without_mods(modified_sequences)
+        data_list = [*masked_values, precursor_charges, mass_mod_sequences, sequences, retention_times, precursor_mz]
+        self.entries = pd.DataFrame(dict(zip(DLIB_COL_NAMES, data_list)))
+
+        # hardcoded entries that we currently not use.
+        # Visit https://bitbucket.org/searleb/encyclopedia/wiki/EncyclopeDIA%20File%20Formats for dlib specs
+        self.entries["Copies"] = 1  # this is hardcorded for now and unused
+        self.entries["Score"] = 0
+        self.entries["CorrelationEncodedLength"] = None
+        self.entries["CorrelationArray"] = None
+        self.entries["RTInSecondsStart"] = None
+        self.entries["RTInSecondsStop"] = None
+        self.entries["MedianChromatogramEncodedLength"] = None
+        self.entries["MedianChromatogramArray"] = None
+        self.entries["SourceFile"] = "Prosit"
+
+        # gather all values for the p2p table and create pandas DataFrame
+        self.p2p = pd.DataFrame({"PeptideSeq": sequences, "isDecoy": False, "ProteinAccession": "unknown"})

spectrum_io/spectral_library/spectral_library.py

@@ -9,25 +9,36 @@
 class SpectralLibrary:
     """Main to initialze a SpectralLibrary obj."""
 
-    def __init__(self, input_dataframe: pd.DataFrame, grpc_dict: dict, output_path: Union[str, Path]):
+    def __init__(
+        self,
+        input_dataframe: pd.DataFrame,
+        grpc_dict: dict,
+        output_path: Union[str, Path],
+        min_intensity_threshold: Optional[float] = 0.05,
+        chunksize: Optional[int] = None,
+    ):
         """
         Initialize a SpectralLibrary obj.
 
         :param input_dataframe: dataframe of sequences, charges, and masses of all library peptides
         :param grpc_dict: GRPC client output dictionary with spectrum, irt, and proteotypicity prediction
         :param output_path: path to output file including file name
+        :param min_intensity_threshold: optional filter for low intensity peaks
+        :param chunksize: optional chunksize for dlib
         """
         if isinstance(output_path, str):
             output_path = Path(output_path)
         self.spectra_input = input_dataframe
         self.grpc_output = grpc_dict
         self.out_path = output_path
+        self.min_intensity_threshold = min_intensity_threshold
+        self.chunksize = chunksize
 
     def load(self):
         """Load predictions from hdf5 file."""
 
     @abstractmethod
-    def write(self, chunksize: Optional[Union[None, int]]):
+    def write(self):
         """Write predictions."""
         pass
 

tests/unit_tests/test_maxquant.py

@@ -62,8 +62,8 @@ def test_update_columns_tmt(self, maxquant_df: pd.DataFrame):
         :param maxquant_df: maxquant df as pd.DataFrame
         """
         prosit_df = mq.MaxQuant.update_columns_for_prosit(maxquant_df, tmt_labeled="tmt")
-        assert prosit_df["MODIFIED_SEQUENCE"][0] == "[UNIMOD:737]DS[UNIMOD:21]DS[UNIMOD:21]WDADAFSVEDPVRK[UNIMOD:737]"
-        assert prosit_df["MODIFIED_SEQUENCE"][3] == "[UNIMOD:737]SS[UNIMOD:21]PTPES[UNIMOD:21]PTMLTK[UNIMOD:737]"
+        assert prosit_df["MODIFIED_SEQUENCE"][0] == "[UNIMOD:737]-DS[UNIMOD:21]DS[UNIMOD:21]WDADAFSVEDPVRK[UNIMOD:737]"
+        assert prosit_df["MODIFIED_SEQUENCE"][3] == "[UNIMOD:737]-SS[UNIMOD:21]PTPES[UNIMOD:21]PTMLTK[UNIMOD:737]"
 
         assert prosit_df["MASS"][0] == 1.0 + 2 * 229.162932
         assert prosit_df["MASS"][3] == 2.0 + 2 * 229.162932
@@ -76,9 +76,9 @@ def test_update_columns_tmt_msa(self, maxquant_df: pd.DataFrame):
         """
         prosit_df = mq.MaxQuant.update_columns_for_prosit(maxquant_df, tmt_labeled="tmt_msa")
         assert (
-            prosit_df["MODIFIED_SEQUENCE_MSA"][0] == "[UNIMOD:737]DS[UNIMOD:23]DS[UNIMOD:23]WDADAFSVEDPVRK[UNIMOD:737]"
+            prosit_df["MODIFIED_SEQUENCE_MSA"][0] == "[UNIMOD:737]-DS[UNIMOD:23]DS[UNIMOD:23]WDADAFSVEDPVRK[UNIMOD:737]"
         )
-        assert prosit_df["MODIFIED_SEQUENCE_MSA"][3] == "[UNIMOD:737]SS[UNIMOD:23]PTPES[UNIMOD:23]PTMLTK[UNIMOD:737]"
+        assert prosit_df["MODIFIED_SEQUENCE_MSA"][3] == "[UNIMOD:737]-SS[UNIMOD:23]PTPES[UNIMOD:23]PTMLTK[UNIMOD:737]"
 
     def test_filter_valid_prosit_sequences(self, invalid_df: pd.DataFrame):
         """Test filter_valid_prosit_sequences."""