0.1.5

output contains gene names instead of STRING IDs

senseppi/__init__.py

-__version__ = "0.1.4"
+__version__ = "0.1.5"
 __author__ = "Konstantin Volzhenin"
 
 from . import model, commands, esm2_model, dataset, utils, network_utils

senseppi/commands/predict_string.py

@@ -62,7 +62,6 @@ def main(params):
     data['preds'] = preds
 
     print(data.sort_values(by=['preds'], ascending=False).to_string())
-    data.to_csv(params.output + '.tsv', sep='\t', index=False)
 
     # Calculate torch metrics based on data['binary_label'] and data['preds']
     torch_labels = torch.tensor(data['binary_label'])
@@ -82,6 +81,12 @@ def main(params):
         string_ids[row['stringId_A']] = row['preferredName_A']
         string_ids[row['stringId_B']] = row['preferredName_B']
 
+    data_to_save = data.copy()
+    data_to_save['seq1'] = data_to_save['seq1'].apply(lambda x: string_ids[x])
+    data_to_save['seq2'] = data_to_save['seq2'].apply(lambda x: string_ids[x])
+    data_to_save = data_to_save.sort_values(by=['preds'], ascending=False)
+    data_to_save.to_csv(params.output + '.tsv', sep='\t', index=False)
+
     # This part was needed to color the pairs belonging to the train data, temporarily removed
 
     # print('Fetching gene names for training set from STRING...')