Merge pull request #128 from wilhelm-lab/feature/internal_to_msp

added new internal_to_msp function

docs/reference.rst

@@ -1,7 +1,7 @@
 How to cite
 ===========
 
-If you want to cite spectrum_fundamentals in your work, please cite the main Oktoberfest publication following:
+If you want to cite spectrum_fundamentals in your work, please cite the main Oktoberfest publication:
 
 .. [1] Picciani M, Gabriel W, Giurcoiu VG et al. (2023),
     *Oktoberfest: Open-source spectral library generation and rescoring pipeline based on Prosit*,

spectrum_fundamentals/mod_string.py

@@ -217,6 +217,41 @@ def internal_to_mod_mass(
     return [regex.sub(replacement_func, seq) for seq in sequences]
 
 
+def internal_to_msp(
+    sequences: Union[List[str], pd.Series],
+    mods: Dict[str, str],
+) -> List[Tuple[str, str]]:
+    """
+    Function to translate an internal modstring to modstring and Mods for MSP format.
+
+    :param sequences: sequences to translate
+    :param mods: dictionary mapping from internal unimod format (keys) to MSP format (values).
+    :return: a tuple for each sequence, containing (Mods, mod_string) for the MSP format
+    """
+    ret_vals = []
+    p = re.compile("|".join(mods.keys()))
+    for seq in sequences:
+        offset = 0
+        mod_list = []
+        matches = p.finditer(seq)
+        for match in matches:
+            replacement = mods[re.escape(match.group())]
+            start, end = match.span()
+            actual_start = start - offset
+            mod_list.append((actual_start, replacement))
+
+            offset += end - start - 1
+
+        mod_string = "; ".join([f"{mod[2:]}@{mod[0]}{pos}" for pos, mod in mod_list])
+        n_mods = len(mod_list)
+        if n_mods > 0:
+            mods_field = f"{n_mods}/{'/'.join([f'{pos},{mod}' for pos, mod in mod_list])}"
+        else:
+            mods_field = "0"
+        ret_vals.append((mods_field, mod_string))
+    return ret_vals
+
+
 def internal_to_mod_names(
     sequences: List[str],
 ) -> List[Tuple[str, str]]: