msreport 0.0.27__py3-none-any.whl → 0.0.29__py3-none-any.whl

msreport/reader.py

@@ -1,4 +1,4 @@
-""" Module for reading result tables from various MS analysis tools and converting them
+"""Module for reading result tables from various MS analysis tools and converting them
 to a standardized format following the MsReport convention.
 
 Currently for MaxQuant and FragPipe protein, peptide, and ion tables are supported, and
@@ -20,19 +20,19 @@ Unified column names:
 - iBAQ intensity "sample name"
 """
 
-from collections import OrderedDict, defaultdict
 import os
-from typing import Any, Callable, Iterable, Optional, Protocol
 import pathlib
 import warnings
+from collections import OrderedDict, defaultdict
+from typing import Any, Callable, Iterable, Optional, Protocol
 
 import numpy as np
 import pandas as pd
 
 import msreport.helper as helper
-from msreport.helper.temp import extract_window_around_position
-from msreport.errors import ProteinsNotInFastaWarning
 import msreport.peptidoform
+from msreport.errors import ProteinsNotInFastaWarning
+from msreport.helper.temp import extract_window_around_position
 
 
 class Protein(Protocol):
@@ -54,6 +54,8 @@ class ProteinDatabase(Protocol):
 class ResultReader:
     """Base Reader class, is by itself not functional."""
 
+    data_directory: str
+    filenames: dict[str, str]
     default_filenames: dict[str, str]
     protected_columns: list[str]
     column_mapping: dict[str, str]
@@ -61,8 +63,8 @@ class ResultReader:
     sample_column_tags: list[str]
 
     def __init__(self):
-        self.data_directory: str = ""
-        self.filenames: dict[str, str] = {}
+        self.data_directory = ""
+        self.filenames = {}
 
     def _read_file(self, which: str, sep: str = "\t") -> pd.DataFrame:
         """Read a result table.
@@ -183,18 +185,16 @@ class MaxQuantReader(ResultReader):
         "MS/MS count",
         "Sequence coverage",
     ]
-    column_mapping: dict[str, str] = dict(
-        [
-            ("Peptides", "Total peptides"),
-            ("Sequence coverage [%]", "Sequence coverage"),
-            ("MS/MS count", "Spectral count Combined"),  # proteinGroups, evidence
-            ("MS/MS Count", "Spectral count Combined"),  # peptides
-            ("Sequence", "Peptide sequence"),  # peptides, evidence
-            ("Sequence length", "Protein length"),
-            ("Mol. weight [kDa]", "Molecular weight [kDa]"),
-            ("Experiment", "Sample"),
-        ]
-    )
+    column_mapping: dict[str, str] = {
+        "Peptides": "Total peptides",
+        "Sequence coverage [%]": "Sequence coverage",
+        "MS/MS count": "Spectral count Combined",  # proteinGroups, evidence
+        "MS/MS Count": "Spectral count Combined",  # peptides
+        "Sequence": "Peptide sequence",  # peptides, evidence
+        "Sequence length": "Protein length",
+        "Mol. weight [kDa]": "Molecular weight [kDa]",
+        "Experiment": "Sample",
+    }
     column_tag_mapping: OrderedDict[str, str] = OrderedDict(
         [("MS/MS count", "Spectral count"), ("iBAQ", "iBAQ intensity")]
     )
@@ -343,7 +343,9 @@ class MaxQuantReader(ResultReader):
         Adds new columns to comply with the MsReport convention. "Modified sequence",
         "Modifications columns", "Modification localization string". "Protein reported
         by software" and "Representative protein", both contain the first entry from
-        "Leading razor protein".
+        "Leading razor protein". "Ion ID" contains unique entries for each ion, which
+        are generated by concatenating the "Modified sequence" and "Charge" columns, and
+        if present, the "Compensation voltage" column.
 
         "Modified sequence" entries contain modifications within square brackets.
         "Modification" entries are strings in the form of "position:modification_tag",
@@ -376,15 +378,19 @@ class MaxQuantReader(ResultReader):
             df["Leading razor protein"]
         )
         df["Representative protein"] = df["Protein reported by software"]
+
         if drop_decoy:
             df = self._drop_decoy(df)
         if rename_columns:
-            df = self._rename_columns(
-                df, True
-            )  # Actually there are no column tags as the table is in long format
+            # Actually there are no column tags as the table is in long format
+            df = self._rename_columns(df, prefix_tag=True)
         if rewrite_modifications and rename_columns:
             df = self._add_peptide_modification_entries(df)
             df = self._add_modification_localization_string(df)
+            df["Ion ID"] = df["Modified sequence"] + "_c" + df["Charge"].astype(str)
+            if "Compensation voltage" in df.columns:
+                _cv = df["Compensation voltage"].astype(str)
+                df["Ion ID"] = df["Ion ID"] + "_cv" + _cv
         return df
 
     def _add_protein_entries(self, df: pd.DataFrame) -> pd.DataFrame:
@@ -576,6 +582,7 @@ class FragPipeReader(ResultReader):
         "peptides": "combined_peptide.tsv",
         "ions": "combined_ion.tsv",
         "ion_evidence": "ion.tsv",
+        "psm_evidence": "psm.tsv",
     }
     isobar_filenames: dict[str, str] = {
         "proteins": "protein.tsv",
@@ -590,20 +597,25 @@ class FragPipeReader(ResultReader):
         "Intensity",
         "MaxLFQ Intensity",
     ]
-    column_mapping: dict[str, str] = dict(
-        [
-            ("Peptide Sequence", "Peptide sequence"),  # Peptide and ion
-            ("Modified Sequence", "Modified sequence"),  # Modified peptide and ion
-            ("Start", "Start position"),  # Peptide and ion
-            ("End", "End position"),  # Peptide and ion
-            ("Combined Total Peptides", "Total peptides"),  # From LFQ
-            ("Total Peptides", "Total peptides"),  # From TMT
-            ("Description", "Protein name"),
-            ("Protein Length", "Protein length"),
-            ("Entry Name", "Protein entry name"),
-            ("Gene", "Gene name"),
-        ]
-    )
+    column_mapping: dict[str, str] = {
+        "Peptide": "Peptide sequence",  # PSM
+        "Modified Peptide": "Modified sequence",  # PSM
+        "Protein Start": "Start position",  # PSM
+        "Protein End": "End position",  # PSM
+        "Number of Missed Cleavages": "Missed cleavage",  # PSM
+        "PeptideProphet Probability": "Probability",  # PSM
+        "Compensation Voltage": "Compensation voltage",  # PSM and ion
+        "Peptide Sequence": "Peptide sequence",  # Peptide and ion
+        "Modified Sequence": "Modified sequence",  # Modified peptide and ion
+        "Start": "Start position",  # Peptide and ion
+        "End": "End position",  # Peptide and ion
+        "Combined Total Peptides": "Total peptides",  # From LFQ
+        "Total Peptides": "Total peptides",  # From TMT
+        "Description": "Protein name",
+        "Protein Length": "Protein length",
+        "Entry Name": "Protein entry name",
+        "Gene": "Gene name",
+    }
     column_tag_mapping: OrderedDict[str, str] = OrderedDict(
         [
             ("MaxLFQ Intensity", "LFQ intensity"),
@@ -743,7 +755,10 @@ class FragPipeReader(ResultReader):
 
         Adds new columns to comply with the MsReport convention. "Modified sequence"
         and "Modifications columns". "Protein reported by software" and "Representative
-        protein", both contain the first entry from "Leading razor protein".
+        protein", both contain the first entry from "Leading razor protein". "Ion ID"
+        contains unique entries for each ion, which are generated by concatenating the
+        "Modified sequence" and "Charge" columns, and if present, the
+        "Compensation voltage" column.
 
         "Modified sequence" entries contain modifications within square brackets.
         "Modification" entries are strings in the form of "position:modification_text",
@@ -783,6 +798,11 @@ class FragPipeReader(ResultReader):
         if rewrite_modifications and rename_columns:
             df = self._add_peptide_modification_entries(df)
             df = self._add_modification_localization_string(df, prefix_column_tags)
+            df["Ion ID"] = df["Modified sequence"] + "_c" + df["Charge"].astype(str)
+            if "Compensation voltage" in df.columns:
+                _cv = df["Compensation voltage"].astype(str)
+                df["Ion ID"] = df["Ion ID"] + "_cv" + _cv
+
         return df
 
     def import_ion_evidence(
@@ -797,7 +817,9 @@ class FragPipeReader(ResultReader):
         Adds new columns to comply with the MsReport convention. "Modified sequence",
         "Modifications", and "Modification localization string" columns. "Protein
         reported by software" and "Representative protein", both contain the first entry
-        from "Leading razor protein".
+        from "Leading razor protein". "Ion ID" contains unique entries for each ion,
+        which are generated by concatenating the "Modified sequence" and "Charge"
+        columns, and if present, the "Compensation voltage" column.
 
         "Modified sequence" entries contain modifications within square brackets.
         "Modification" entries are strings in the form of "position:modification_text",
@@ -850,6 +872,9 @@ class FragPipeReader(ResultReader):
         df = pd.concat(ion_tables, ignore_index=True)
 
         # --- Process dataframe --- #
+        df["Ion ID"] = df["Modified Sequence"] + "_c" + df["Charge"].astype(str)
+        if "Compensation Voltage" in df.columns:
+            df["Ion ID"] = df["Ion ID"] + "_cv" + df["Compensation Voltage"].astype(str)
         # FUTURE: replace this by _add_protein_entries(df, False) if FragPipe adds
         #         'Indistinguishable Proteins' to the ion table.
         df["Protein reported by software"] = _extract_protein_ids(df["Protein"])
@@ -861,6 +886,76 @@ class FragPipeReader(ResultReader):
             df = self._add_modification_localization_string(df, prefix_column_tags)
         return df
 
+    def import_psm_evidence(
+        self,
+        filename: Optional[str] = None,
+        rename_columns: bool = True,
+        rewrite_modifications: bool = True,
+    ):
+        """Concatenate all "psm.tsv" files and return a processed dataframe.
+
+        Args:
+            filename: Allows specifying an alternative filename, otherwise the default
+                filename is used.
+            rename_columns: If True, columns are renamed according to the MsReport
+                convention; default True.
+            rewrite_modifications: If True, the peptide format in "Modified sequence" is
+                changed according to the MsReport convention, and a "Modifications" is
+                added to contains the amino acid position for all modifications.
+                Requires 'rename_columns' to be true. Default True.
+
+        Returns:
+            A DataFrame containing the processed psm evidence tables.
+        """
+        if filename is None:
+            filename = self.default_filenames["psm_evidence"]
+
+        psm_table_paths = []
+        for path in pathlib.Path(self.data_directory).iterdir():
+            psm_table_path = path / filename
+            if path.is_dir() and psm_table_path.exists():
+                psm_table_paths.append(psm_table_path)
+
+        psm_tables = []
+        for filepath in psm_table_paths:
+            table = pd.read_csv(filepath, sep="\t", low_memory=False)
+            str_cols = table.select_dtypes(include=["object"]).columns
+            table.loc[:, str_cols] = table.loc[:, str_cols].fillna("")
+
+            table["Sample"] = filepath.parent.name
+            psm_tables.append(table)
+        df = pd.concat(psm_tables, ignore_index=True)
+
+        df["Protein reported by software"] = _extract_protein_ids(df["Protein"])
+        df["Representative protein"] = df["Protein reported by software"]
+        df["Mapped Proteins"] = df["Mapped Proteins"].astype(str).replace("nan", "")
+
+        # FP only lists additional mapped proteins in the "Mapped Proteins" column
+        # MsReport reports all matching proteins in the "Mapped proteins" column
+        mapped_proteins_entries = []
+        for protein, mapped_protein_fp in zip(
+            df["Representative protein"], df["Mapped Proteins"], strict=True
+        ):
+            if mapped_protein_fp == "":
+                mapped_proteins = [protein]
+            else:
+                additional_mapped_proteins = msreport.reader._extract_protein_ids(
+                    mapped_protein_fp.split(", ")
+                )
+                mapped_proteins = [protein] + additional_mapped_proteins
+            mapped_proteins_entries.append(";".join(mapped_proteins))
+        df["Mapped proteins"] = mapped_proteins_entries
+
+        if rename_columns:
+            df = self._rename_columns(df, prefix_tag=True)
+        if rewrite_modifications and rename_columns:
+            mod_entries = _generate_modification_entries_from_assigned_modifications(
+                df["Peptide sequence"], df["Assigned Modifications"]
+            )
+            df["Modified sequence"] = mod_entries["Modified sequence"]
+            df["Modifications"] = mod_entries["Modifications"]
+        return df
+
     def _add_protein_entries(self, df: pd.DataFrame) -> pd.DataFrame:
         """Adds standardized protein entry columns to the data frame.
 
@@ -1038,40 +1133,32 @@ class SpectronautReader(ResultReader):
         "design": "conditionsetup",
     }
     protected_columns: list[str] = []
-    column_mapping: dict[str, str] = dict(
-        [
-            ("R.FileName", "Filename"),
-            ("R.Label", "Sample"),
-            ("PG.Qvalue", "Protein qvalue"),
-            ("PG.Cscore", "Protein cscore"),
-            ("PG.NrOfStrippedSequencesIdentified (Experiment-wide)", "Total peptides"),
-            ("PG.NrOfPrecursorsIdentified (Experiment-wide)", "Total ions"),
-            ("PG.Cscore", "Cscore"),
-            ("PEP.StrippedSequence", "Peptide sequence"),
-            ("PEP.AllOccurringProteinAccessions", "Mapped proteins"),
-            ("EG.ModifiedSequence", "Modified sequence"),
-            ("EG.CompensationVoltage", "Compensation voltage"),
-            ("EG.Qvalue", "Qvalue"),
-            ("EG.ApexRT", "Apex retention time"),
-            ("EG.DatapointsPerPeak", "Datapoints per peak"),
-            ("EG.FWHM", "FWHM"),
-            ("EG.SignalToNoise", "Signal to noise"),
-            ("FG.FragmentCount", "Fragment count"),
-            ("FG.Charge", "Charge"),
-            ("FG.MS1Quantity", "MS1 intensity"),
-            ("FG.MS1RawQuantity", "MS1 raw intensity"),
-            ("FG.MS2Quantity", "MS2 intensity"),
-            ("FG.MS2RawQuantity", "MS2 raw intensity"),
-            ("FG.MeasuredMz", "Observed m/z"),
-            ("FG.TheoreticalMz", "Theoretical m/z"),
-            ("FG.CalibratedMz", "Calibrated m/z"),
-            # ("PG.ProteinAccessions", ""),
-            # ("EG.HasLocalizationInformation", ""),
-            # ("EG.PTMLocalizationProbabilities", ""),
-            # ("EG.UsedForProteinGroupQuantity", ""),
-            # Modified peptides need to be parsed and rewritten
-        ]
-    )
+    column_mapping: dict[str, str] = {
+        "R.FileName": "Filename",
+        "R.Label": "Sample",
+        "PG.Qvalue": "Protein qvalue",
+        "PG.Cscore": "Protein cscore",
+        "PG.NrOfStrippedSequencesIdentified (Experiment-wide)": "Total peptides",
+        "PG.NrOfPrecursorsIdentified (Experiment-wide)": "Total ions",
+        "PEP.StrippedSequence": "Peptide sequence",
+        "PEP.AllOccurringProteinAccessions": "Mapped proteins",
+        "EG.ModifiedSequence": "Modified sequence",
+        "EG.CompensationVoltage": "Compensation voltage",
+        "EG.Qvalue": "Qvalue",
+        "EG.ApexRT": "Apex retention time",
+        "EG.DatapointsPerPeak": "Datapoints per peak",
+        "EG.FWHM": "FWHM",
+        "EG.SignalToNoise": "Signal to noise",
+        "FG.FragmentCount": "Fragment count",
+        "FG.Charge": "Charge",
+        "FG.MS1Quantity": "MS1 intensity",
+        "FG.MS1RawQuantity": "MS1 raw intensity",
+        "FG.MS2Quantity": "MS2 intensity",
+        "FG.MS2RawQuantity": "MS2 raw intensity",
+        "FG.MeasuredMz": "Observed m/z",
+        "FG.TheoreticalMz": "Theoretical m/z",
+        "FG.CalibratedMz": "Calibrated m/z",
+    }
     sample_column_tags: list[str] = [
         ".PG.NrOfPrecursorsIdentified",
         ".PG.IBAQ",
@@ -1324,12 +1411,14 @@ class SpectronautReader(ResultReader):
         filename: Optional[str] = None,
         filetag: Optional[str] = None,
         rename_columns: bool = True,
-    ) -> None:
+    ) -> pd.DataFrame:
         """Reads an ion evidence file (long format) and returns a processed dataframe.
 
         Adds new columns to comply with the MsReport convention. "Protein reported
         by software" and "Representative protein", both contain the first entry from
-        "PG.ProteinAccessions".
+        "PG.ProteinAccessions". "Ion ID" contains unique entries for each ion, which are
+        generated by concatenating the "Modified sequence" and "Charge" columns, and if
+        present, the "Compensation voltage" column.
 
         (!) Note that the modified sequence and modification localization probabilities
         are currently not processed.
@@ -1367,6 +1456,11 @@ class SpectronautReader(ResultReader):
         df = self._add_protein_entries(df)
         if rename_columns:
             df = self._rename_columns(df, True)
+            df["Ion ID"] = df["Modified sequence"] + "_c" + df["Charge"].astype(str)
+            if "Compensation voltage" in df.columns:
+                _cv = df["Compensation voltage"].astype(str)
+                df["Ion ID"] = df["Ion ID"] + "_cv" + _cv
+
         return df
 
     def _tidy_up_sample_columns(self, df: pd.DataFrame) -> pd.DataFrame:
@@ -1462,7 +1556,7 @@ def sort_leading_proteins(
     if penalize_contaminants is not None:
         contaminant_encoding = {"False": 0, "True": 1, False: 0, True: 1}
 
-    for idx, row in table.iterrows():
+    for _, row in table.iterrows():
         protein_ids = row["Leading proteins"].split(";")
 
         sorting_info = [[] for _ in protein_ids]
@@ -1559,6 +1653,7 @@ def add_protein_annotation(
         warnings.warn(
             f"Some proteins could not be annotated: {repr(proteins_not_in_db)}",
             ProteinsNotInFastaWarning,
+            stacklevel=2,
         )
 
     annotations = {}
@@ -1636,9 +1731,10 @@ def add_protein_site_annotation(
         warnings.warn(
             f"Some proteins could not be annotated: {repr(proteins_not_in_db)}",
             ProteinsNotInFastaWarning,
+            stacklevel=2,
         )
 
-    annotations = {
+    annotations: dict[str, list[str]] = {
         "Modified residue": [],
         "Sequence window": [],
     }
@@ -1702,6 +1798,7 @@ def add_leading_proteins_annotation(
         warnings.warn(
             f"Some proteins could not be annotated: {repr(proteins_not_in_db)}",
             ProteinsNotInFastaWarning,
+            stacklevel=2,
         )
 
     annotations = {}
@@ -1853,7 +1950,7 @@ def add_peptide_positions(
             find matching entries in the FASTA files.
     """
     # not tested #
-    peptide_positions = {"Start position": [], "End position": []}
+    peptide_positions: dict[str, list[int]] = {"Start position": [], "End position": []}
     proteins_not_in_db = []
     for peptide, protein_id in zip(table[peptide_column], table[protein_column]):
         if protein_id in protein_db:
@@ -1875,6 +1972,7 @@ def add_peptide_positions(
         warnings.warn(
             f"Some peptides could not be annotated: {repr(proteins_not_in_db)}",
             ProteinsNotInFastaWarning,
+            stacklevel=2,
         )
 
 
@@ -1894,10 +1992,10 @@ def add_protein_modifications(table: pd.DataFrame):
             for peptide_site, mod in [m.split(":") for m in mod_entry.split(";")]:
                 protein_site = int(peptide_site) + start_pos - 1
                 protein_mods.append([str(protein_site), mod])
-            protein_mods = ";".join([f"{pos}:{mod}" for pos, mod in protein_mods])
+            protein_mod_string = ";".join([f"{pos}:{mod}" for pos, mod in protein_mods])
         else:
-            protein_mods = ""
-        protein_modification_entries.append(protein_mods)
+            protein_mod_string = ""
+        protein_modification_entries.append(protein_mod_string)
     table["Protein modifications"] = protein_modification_entries
 
 
@@ -2074,7 +2172,7 @@ def _process_protein_entries(
         A dataframe containing the columns "Protein reported by software",
         "Leading proteins", "Representative protein", and "Potential contaminant".
     """
-    new_entries = {
+    new_entries: dict[str, list[str | bool]] = {
         "Protein reported by software": [],
         "Representative protein": [],
         "Potential contaminant": [],
@@ -2147,6 +2245,57 @@ def _generate_modification_entries(
     return entries
 
 
+def _generate_modification_entries_from_assigned_modifications(
+    sequences: Iterable[str],
+    assigned_modifications: Iterable[str],
+) -> dict[str, list[str]]:
+    modified_sequence_entries = []
+    modification_entries = []
+    for sequence, modifications_entry in zip(sequences, assigned_modifications):
+        modifications = _extract_fragpipe_assigned_modifications(
+            modifications_entry, sequence
+        )
+        modified_sequence = helper.modify_peptide(sequence, modifications)
+        modification_entry = ";".join([f"{pos}:{mod}" for pos, mod in modifications])
+        modified_sequence_entries.append(modified_sequence)
+        modification_entries.append(modification_entry)
+
+    entries = {
+        "Modified sequence": modified_sequence_entries,
+        "Modifications": modification_entries,
+    }
+    return entries
+
+
+def _extract_fragpipe_assigned_modifications(
+    modifications_entry: str,
+    sequence: str,
+) -> list[tuple[int, str]]:
+    """Extracts modifications from a FragPipe "Modifications" entry.
+
+    Example for a modification entry: "N-term(42.0106),8C(57.0215)"
+
+    Returns:
+        A list of tuples, where each tuple contains the position of the modification and
+        the modification text. The position is one-indexed, meaning that the first amino
+        acid position is 1. N-term and C-term are represented as 0 and len(sequence)
+        respectively.
+    """
+    if modifications_entry == "":
+        return []
+    modifications = []
+    for mod_entry in modifications_entry.split(","):
+        position_entry, modification = mod_entry.split(")")[0].split("(")
+        if position_entry == "N-term":
+            position = 0
+        elif position_entry == "C-term":
+            position = len(sequence)
+        else:
+            position = int(position_entry[:-1])
+        modifications.append((position, modification))
+    return modifications
+
+
 def extract_maxquant_localization_probabilities(localization_entry: str) -> dict:
     """Extract localization probabilites from a MaxQuant "Probabilities" entry.
 
@@ -2189,7 +2338,7 @@ def extract_fragpipe_localization_probabilities(localization_entry: str) -> dict
     ... )
     {'15.9949': {3: 1.0}, '79.9663': {4: 0.334, 6: 0.666}}
     """
-    modification_probabilities = {}
+    modification_probabilities: dict[str, dict[int, float]] = {}
     for modification_entry in filter(None, localization_entry.split(";")):
         specified_modification, probability_sequence = modification_entry.split("@")
         _, modification = specified_modification.split(":")
@@ -2247,7 +2396,7 @@ def _create_protein_annotations_from_db(
     protein_db: ProteinDatabase,
     query_function: Callable,
     default_value: Any,
-) -> list[str]:
+) -> list[Any]:
     """Returns a list of multi protein entry annotations.
 
     Used to generate protein annotations for protein entries. For each protein id an
@@ -2274,9 +2423,9 @@ def _create_protein_annotations_from_db(
         if protein_id in protein_db:
             db_entry = protein_db[protein_id]
             query_result = query_function(db_entry, default_value)
+            annotation_values.append(query_result)
         else:
-            query_result = default_value
-        annotation_values.append(query_result)
+            annotation_values.append(default_value)
     return annotation_values
 
 

msreport/rinterface/__init__.py

@@ -1,3 +1,16 @@
-""" Python interface to custome R scripts. """
-from .limma import multi_group_limma, two_group_limma
-from .rinstaller import r_package_version
+"""Python interface to custome R scripts."""
+
+from msreport.errors import OptionalDependencyError
+
+try:
+    from .limma import multi_group_limma, two_group_limma
+    from .rinstaller import r_package_version
+except ImportError as err:
+    raise OptionalDependencyError(
+        "R integration is not available. R must be installed and configured before "
+        "installing optional R dependencies using 'pip install msreport[R]'. For "
+        "more information, see: https://github.com/hollenstein/msreport"
+    ) from err
+
+
+__all__ = ["multi_group_limma", "two_group_limma", "r_package_version"]

msreport/rinterface/limma.py

@@ -1,4 +1,5 @@
-""" Python interface to custome R scripts. """
+"""Python interface to custome R scripts."""
+
 import os
 
 import pandas as pd

msreport/rinterface/rinstaller.py

@@ -1,9 +1,9 @@
-from rpy2.robjects.packages import importr
-import rpy2.robjects.packages as rpackages
 import rpy2.robjects as robjects
+import rpy2.robjects.packages as rpackages
+from rpy2.robjects.packages import importr
 
 
-def r_package_version(package_name: str) -> (str, str):
+def r_package_version(package_name: str) -> str:
     """Returns the version number of an installed R package."""
     with robjects.conversion.localconverter(robjects.default_converter):
         utils = importr("utils")