Merge pull request #231 from MannLabs/add_psm_reader_tests

Add psm reader tests

alphabase/psm_reader/msfragger_reader.py

@@ -137,7 +137,7 @@ def _translate_decoy(self, origin_df=None):
         )
 
         self._psm_df.proteins = self._psm_df.proteins.apply(lambda x: ";".join(x))
-        if not self.keep_decoy:
+        if not self._keep_decoy:
             self._psm_df["to_remove"] += self._psm_df.decoy > 0
 
     def _translate_score(self, origin_df=None):

alphabase/psm_reader/psm_reader.py

@@ -49,7 +49,7 @@ def translate_other_modification(mod_str: str, mod_dict: dict) -> str:
         return ";".join(ret_mods), []
 
 
-def keep_modifications(mod_str: str, mod_set: set) -> str:
+def _keep_modifications(mod_str: str, mod_set: set) -> str:
     """
     Check if modifications of `mod_str` are in `mod_set`.
 
@@ -162,8 +162,8 @@ def __init__(
             self._init_column_mapping()
 
         self._psm_df = pd.DataFrame()
-        self.keep_fdr = fdr
-        self.keep_decoy = keep_decoy
+        self._keep_fdr = fdr
+        self._keep_decoy = keep_decoy
         self._min_max_rt_norm = False
         self._engine_rt_unit = rt_unit
         self._min_irt_value = -100
@@ -311,7 +311,7 @@ def _translate_score(self, origin_df: pd.DataFrame = None):
     def _get_table_delimiter(self, _filename):
         return get_delimiter(_filename)
 
-    def normalize_rt(self):
+    def _normalize_rt(self):
         if "rt" in self.psm_df.columns:
             if self._engine_rt_unit == "second":
                 # self.psm_df['rt_sec'] = self.psm_df.rt
@@ -336,14 +336,11 @@ def normalize_rt(self):
                 (self.psm_df.rt - min_rt) / (max_rt - min_rt)
             ).clip(0, 1)
 
-    def norm_rt(self):
-        self.normalize_rt()
-
     def normalize_rt_by_raw_name(self):
         if "rt" not in self.psm_df.columns:
             return
         if "rt_norm" not in self.psm_df.columns:
-            self.norm_rt()
+            self._normalize_rt()
         if "raw_name" not in self.psm_df.columns:
             return
         for _, df_group in self.psm_df.groupby("raw_name"):
@@ -490,8 +487,8 @@ def _post_process(self, origin_df: pd.DataFrame):
 
         keep_rows = np.ones(len(self._psm_df), dtype=bool)
         if "fdr" in self._psm_df.columns:
-            keep_rows &= self._psm_df.fdr <= self.keep_fdr
-        if "decoy" in self._psm_df.columns and not self.keep_decoy:
+            keep_rows &= self._psm_df.fdr <= self._keep_fdr
+        if "decoy" in self._psm_df.columns and not self._keep_decoy:
             keep_rows &= self._psm_df.decoy == 0
 
         self._psm_df = self._psm_df[keep_rows]
@@ -528,7 +525,7 @@ def filter_psm_by_modifications(
                 ]
             )
         self._psm_df.mods = self._psm_df.mods.apply(
-            keep_modifications, mod_set=include_mod_set
+            _keep_modifications, mod_set=include_mod_set
         )
 
         self._psm_df.dropna(subset=["mods"], inplace=True)

alphabase/psm_reader/sage_reader.py

@@ -606,12 +606,12 @@ def _transform_table(self, origin_df):
         self.psm_df.drop(columns=["scannr"], inplace=True)
 
     def _translate_decoy(self, origin_df):
-        if not self.keep_decoy:
+        if not self._keep_decoy:
             self._psm_df = self.psm_df[~self.psm_df["decoy"]]
 
-        self._psm_df = self.psm_df[self.psm_df["fdr"] <= self.keep_fdr]
-        self._psm_df = self.psm_df[self.psm_df["peptide_fdr"] <= self.keep_fdr]
-        self._psm_df = self.psm_df[self.psm_df["protein_fdr"] <= self.keep_fdr]
+        self._psm_df = self.psm_df[self.psm_df["fdr"] <= self._keep_fdr]
+        self._psm_df = self.psm_df[self.psm_df["peptide_fdr"] <= self._keep_fdr]
+        self._psm_df = self.psm_df[self.psm_df["protein_fdr"] <= self._keep_fdr]
 
         # drop peptide_fdr, protein_fdr
         self._psm_df.drop(columns=["peptide_fdr", "protein_fdr"], inplace=True)

alphabase/spectral_library/reader.py

@@ -170,7 +170,7 @@ def _get_fragment_intensity(self, lib_df: pd.DataFrame):
         ]
 
         # by default, all non-fragment columns are used to group the library
-        non_fragment_columns = list(set(lib_df.columns) - set(fragment_columns))
+        non_fragment_columns = sorted(list(set(lib_df.columns) - set(fragment_columns)))
 
         for keys, df_group in tqdm(lib_df.groupby(non_fragment_columns)):
             precursor_columns = dict(zip(non_fragment_columns, keys))

nbs_tests/psm_reader/psm_reader.ipynb

@@ -32,9 +32,7 @@
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": [
-    "from alphabase.psm_reader.psm_reader import *"
-   ]
+   "source": "from alphabase.psm_reader.psm_reader import translate_other_modification, PSMReaderBase, psm_reader_yaml"
   },
   {
    "cell_type": "code",
@@ -43,11 +41,11 @@
    "outputs": [],
    "source": [
     "#| hide\n",
-    "assert 'a',[] == translate_other_modification('A', {'A':'a','B':'b'})\n",
-    "assert 'b',[] == translate_other_modification('B', {'A':'a','B':'b'})\n",
-    "assert 'a;a',[] == translate_other_modification('A;A', {'A':'a','B':'b'})\n",
-    "assert 'a;b',[] == translate_other_modification('A;B', {'A':'a','B':'b'})\n",
-    "assert 'a;b',['X'] == translate_other_modification('A;B;X', {'A':'a','B':'b'})"
+    "assert 'a', [] == translate_other_modification('A', {'A': 'a', 'B': 'b'})\n",
+    "assert 'b', [] == translate_other_modification('B', {'A': 'a', 'B': 'b'})\n",
+    "assert 'a;a', [] == translate_other_modification('A;A', {'A': 'a', 'B': 'b'})\n",
+    "assert 'a;b', [] == translate_other_modification('A;B', {'A': 'a', 'B': 'b'})\n",
+    "assert 'a;b', ['X'] == translate_other_modification('A;B;X', {'A': 'a', 'B': 'b'})"
    ]
   },
   {
@@ -106,34 +104,34 @@
     "  'Acetyl@Protein_N-term': [\n",
     "    '_(Acetyl (Protein_N-term))',\n",
     "    '_(ac)',\n",
-    "  ]\n",
+    "  ],\n",
     "  'Carbamidomethyl@C': [\n",
     "    'C(Carbamidomethyl (C))',\n",
-    "  ]\n",
+    "  ],\n",
     "  'Oxidation@M': [\n",
     "    'M(Oxidation (M))',\n",
     "    'M(ox)',\n",
-    "  ]\n",
+    "  ],\n",
     "  'Phospho@S': [\n",
     "    'S(Phospho (S))',\n",
     "    'S(Phospho (ST))',\n",
     "    'S(Phospho (STY))',\n",
     "    'S(ph)',\n",
     "    'pS',\n",
-    "  ]\n",
+    "  ],\n",
     "  'Phospho@T': [\n",
     "    'T(Phospho (T))',\n",
     "    'T(Phospho (ST))',\n",
     "    'T(Phospho (STY))',\n",
     "    'T(ph)',\n",
     "    'pT',\n",
-    "  ]\n",
+    "  ],\n",
     "  'Phospho@Y': [\n",
     "    'Y(Phospho (Y))',\n",
     "    'Y(Phospho (STY))',\n",
     "    'Y(ph)',\n",
     "    'pY',\n",
-    "  ]\n",
+    "  ],\n",
     "  'Deamidated@N': ['N(Deamidation (NQ))','N(de)']\n",
     "  'Deamidated@Q': ['Q(Deamidation (NQ))','Q(de)']\n",
     "  'GlyGly@K': ['K(GlyGly (K))', 'K(gl)']\n",