common code to finalize tables

tiger.py

@@ -302,9 +302,6 @@ def predict_off_target(off_targets: pd.DataFrame, model: tf.keras.Model):
     lfc_estimate = model.predict(model_inputs, batch_size=BATCH_SIZE_COMPUTE, verbose=False)
     off_targets[SCORE_COL] = prediction_transform(tf.squeeze(lfc_estimate).numpy())
 
-    # trim context sequence
-    off_targets[TARGET_COL] = off_targets[TARGET_COL].apply(lambda seq: seq[CONTEXT_5P:len(seq) - CONTEXT_3P])
-
     return off_targets.reset_index(drop=True)
 
 
@@ -345,10 +342,11 @@ def tiger_exhibit(transcripts: pd.DataFrame, mode: str, check_off_targets: bool,
         off_target_predictions = predict_off_target(off_target_candidates, model=tiger)
         off_target_predictions = off_target_predictions.sort_values(SCORE_COL, ascending=False)
 
-    # reverse guide sequences
-    on_target_predictions[GUIDE_COL] = on_target_predictions[GUIDE_COL].apply(lambda s: s[::-1])
-    if check_off_targets and len(off_target_predictions) > 0:
-        off_target_predictions[GUIDE_COL] = off_target_predictions[GUIDE_COL].apply(lambda s: s[::-1])
+    # finalize tables
+    for df in [on_target_predictions, titration_predictions, off_target_predictions]:
+        if df is not None:
+            df[GUIDE_COL] = df[GUIDE_COL].apply(lambda s: s[::-1])  # reverse guide sequences
+            df[TARGET_COL] = df[TARGET_COL].apply(lambda seq: seq[CONTEXT_5P:len(seq) - CONTEXT_3P])  # remove context
 
     return on_target_predictions, titration_predictions, off_target_predictions