FIX smaller bug fixes

alphadia/data/thermo.py

@@ -1,6 +1,8 @@
 # native imports
 import math
 import os
+import logging
+logger = logging.getLogger()
 
 # alphadia imports
 from alphadia import utils
@@ -114,6 +116,7 @@ def __init__(
             self, 
             path, 
             astral_ms1=False,
+            cv=None,
             **kwargs
             ):
             super().__init__(**kwargs)
@@ -122,6 +125,7 @@ def __init__(
             self.sample_name = os.path.basename(self.raw_file_path)
 
             self.astral_ms1 = astral_ms1
+            self.cv = cv
             self.filter_spectra()
 
             self.cycle = calculate_cycle(self.spectrum_df)
@@ -145,16 +149,29 @@ def __init__(
             self.frame_max_index = len(self.rt_values)-1
 
         def filter_spectra(self):
+
+            print(self.cv, 'cv' in self.spectrum_df.columns)
+
+            # filter for astral MS1
             if self.astral_ms1:
                 self.spectrum_df = self.spectrum_df[self.spectrum_df['nce'] > 0.1]
                 self.spectrum_df.loc[self.spectrum_df['nce'] < 1.1, 'ms_level'] = 1
                 self.spectrum_df.loc[self.spectrum_df['nce'] < 1.1, 'precursor_mz'] = -1.0
                 self.spectrum_df.loc[self.spectrum_df['nce'] < 1.1, 'isolation_lower_mz'] = -1.0
                 self.spectrum_df.loc[self.spectrum_df['nce'] < 1.1, 'isolation_upper_mz'] = -1.0
-                self.spectrum_df['spec_idx'] = np.arange(len(self.spectrum_df))
             else:
                 self.spectrum_df = self.spectrum_df[(self.spectrum_df['nce'] < 0.1) | (self.spectrum_df['nce'] > 1.1)]
-                self.spectrum_df['spec_idx'] = np.arange(len(self.spectrum_df))
+
+            # filter for cv
+            if self.cv is not None:
+                if 'cv' in self.spectrum_df.columns:
+                    # use np.isclose to account for floating point errors
+                    logger.info(f"Filtering for CV {self.cv}")
+                    logger.info(f"Before: {len(self.spectrum_df)}")
+                    self.spectrum_df = self.spectrum_df[np.isclose(self.spectrum_df['cv'], self.cv, atol=0.1)]
+                    logger.info(f"After: {len(self.spectrum_df)}")
+
+            self.spectrum_df['spec_idx'] = np.arange(len(self.spectrum_df))
 
         def jitclass(self):
             return ThermoJIT(

alphadia/hybridselection.py

@@ -1591,7 +1591,7 @@ def plot_candidates(
     has_mobility = height_px > 2
 
     if has_mobility:
-        fig_size = (max(width_px/500 * 8,5), height_px/500 * 5 )
+        fig_size = (max(width_px/500 * 8,5), height_px/100 * 5 )
         gridspec_kw = {'height_ratios': [1,9]}
     else:
         fig_size = (max(width_px/500 * 8,5), 5)

alphadia/libtransform.py

@@ -303,6 +303,15 @@ def forward(self, input: SpecLibBase) -> SpecLibBase:
         if len(input.precursor_df) == 0:
             logger.warning(f'Input library has no precursor information. Skipping RT normalization')
             return input
+
+        if 'rt' not in input.precursor_df.columns:
+            if 'rt_norm' in input.precursor_df.columns:
+                logger.warning(f'Input library already contains normalized RT information. Skipping RT normalization')
+                return input
+            else:
+                logger.warning(f'Input library has no RT information. Skipping RT normalization')
+                return input
+
         percentiles = np.percentile(input.precursor_df['rt'], [0.1,99.9])
         input._precursor_df['rt'] = np.clip(input._precursor_df['rt'], percentiles[0], percentiles[1])
 

alphadia/planning.py

@@ -216,6 +216,10 @@ def run(self,
                 del workflow
 
             except Exception as e:
+                # get full traceback
+                import traceback
+                traceback.print_exc()
+
                 print(e)
                 logger.error(f'Workflow failed for {raw_name} with error {e}')
                 continue

alphadia/workflow/base.py

@@ -175,10 +175,17 @@ def _get_dia_data_object(
             )
         elif file_extension == '.raw':
             self.reporter.log_metric('raw_data_type', 'thermo')
+            # check if cv selection exists
+            cv = None
+            if 'raw_data_loading' in self.config:
+                if 'cv' in self.config['raw_data_loading']:
+                    cv = self.config['raw_data_loading']['cv']
+
             return thermo.Thermo(
                 dia_data_path,
                 astral_ms1= self.config['general']['astral_ms1'],
                 process_count = self.config['general']['thread_count'],
+                cv=cv,
             )
         else:
             raise ValueError(f'Unknown file extension {file_extension} for file at {dia_data_path}')