Added tmt modifications for msfragger and mascot

spectrum_io/search_result/filter.py

@@ -2,6 +2,7 @@
 import re
 
 import pandas as pd
+import spectrum_fundamentals.constants as c
 
 logger = logging.getLogger(__name__)
 
@@ -27,3 +28,17 @@ def filter_valid_prosit_sequences(df: pd.DataFrame) -> pd.DataFrame:
     logger.info(f"#sequences after filtering for valid prosit sequences: {len(df.index)}")
 
     return df
+
+
+def add_tmt_mod(mass: float, seq: str, unimod_tag: str) -> float:
+    """
+    Add tmt modification.
+
+    :param mass: mass without tmt modification
+    :param seq: sequence of the peptide
+    :param unimod_tag: UNIMOD tag for the modification
+    :return: mass as float
+    """
+    num_of_tmt = seq.count(unimod_tag)
+    mass += num_of_tmt * c.MOD_MASSES[f"{unimod_tag}"]
+    return mass

spectrum_io/search_result/mascot.py

@@ -7,7 +7,7 @@
 import spectrum_fundamentals.constants as c
 from spectrum_fundamentals.mod_string import internal_without_mods
 
-from .filter import filter_valid_prosit_sequences
+from .filter import add_tmt_mod, filter_valid_prosit_sequences
 
 logger = logging.getLogger(__name__)
 
@@ -92,6 +92,13 @@ def read_mascot(path: Union[str, Path], tmt_labeled: str) -> pd.DataFrame:
 
     df["MODIFIED_SEQUENCE"] = sequences
 
+    if tmt_labeled != "":
+        unimod_tag = c.TMT_MODS[tmt_labeled]
+        logger.info("Adding TMT fixed modifications")
+        df["MODIFIED_SEQUENCE"] = df["MODIFIED_SEQUENCE"].str.replace("K", f"K{unimod_tag}")
+        df["MODIFIED_SEQUENCE"] = unimod_tag + "-" + df["MODIFIED_SEQUENCE"]
+        df["MASS"] = df.apply(lambda x: add_tmt_mod(x.MASS, x.MODIFIED_SEQUENCE, unimod_tag), axis=1)
+
     df["SEQUENCE"] = internal_without_mods(df["MODIFIED_SEQUENCE"])
     df["PEPTIDE_LENGTH"] = df["SEQUENCE"].apply(lambda x: len(x))
 

spectrum_io/search_result/maxquant.py

@@ -6,7 +6,7 @@
 import spectrum_fundamentals.constants as c
 from spectrum_fundamentals.mod_string import internal_without_mods, maxquant_to_internal
 
-from .filter import filter_valid_prosit_sequences
+from .filter import add_tmt_mod, filter_valid_prosit_sequences
 
 logger = logging.getLogger(__name__)
 
@@ -88,17 +88,3 @@ def update_columns_for_prosit(df: pd.DataFrame, tmt_labeled: str) -> pd.DataFram
     df["PEPTIDE_LENGTH"] = df["SEQUENCE"].apply(lambda x: len(x))
 
     return df
-
-
-def add_tmt_mod(mass: float, seq: str, unimod_tag: str) -> float:
-    """
-    Add tmt modification.
-
-    :param mass: mass without tmt modification
-    :param seq: sequence of the peptide
-    :param unimod_tag: UNIMOD tag for the modification
-    :return: mass as float
-    """
-    num_of_tmt = seq.count(unimod_tag)
-    mass += num_of_tmt * c.MOD_MASSES[f"{unimod_tag}"]
-    return mass

spectrum_io/search_result/msfragger.py

@@ -8,7 +8,7 @@
 from spectrum_fundamentals.mod_string import internal_without_mods
 from tqdm import tqdm
 
-from .filter import filter_valid_prosit_sequences
+from .filter import add_tmt_mod, filter_valid_prosit_sequences
 
 logger = logging.getLogger(__name__)
 
@@ -38,7 +38,7 @@ def read_msfragger(path: Union[str, Path], tmt_labeled: str) -> pd.DataFrame:
 
     df = pd.concat(ms_frag_results)
 
-    df = update_columns_for_prosit(df, "")
+    df = update_columns_for_prosit(df, tmt_labeled)
     return filter_valid_prosit_sequences(df)
 
 
@@ -55,6 +55,11 @@ def update_columns_for_prosit(df, tmt_labeled: str) -> pd.DataFrame:
     df["MASS"] = df["precursor_neutral_mass"]
     df["PEPTIDE_LENGTH"] = df["peptide"].apply(lambda x: len(x))
     df["MODIFIED_SEQUENCE"] = msfragger_to_internal(df["modified_peptide"])
+    if tmt_labeled != "":
+        unimod_tag = c.TMT_MODS[tmt_labeled]
+        df["MODIFIED_SEQUENCE"] = df["MODIFIED_SEQUENCE"].str.replace("K", f"K{unimod_tag}")
+        df["MODIFIED_SEQUENCE"] = unimod_tag + "-" + df["MODIFIED_SEQUENCE"]
+        df["MASS"] = df.apply(lambda x: add_tmt_mod(x.MASS, x.MODIFIED_SEQUENCE, unimod_tag), axis=1)
     df.rename(
         columns={
             "assumed_charge": "PRECURSOR_CHARGE",