getting scores by transcript

src/main.py

@@ -11,23 +11,10 @@ def read_root():
     return {"Hello": "World"}
 
 
-@app.get("/items/{item_id}")
-def read_item(item_id: int, q: Union[str, None] = None):
-    return {"item_id": item_id, "q": q}
-
-
-@app.get("/loc/{chromosome}/{start}/{end}")
-def get_scores(chromosome: str, start: int, end: int):
-    # TODO create interface to tabix to query range for DITTO scores
-    return {"loc": f"{chromosome}:{start}-{end}"}
-
-
-@app.get("/var/{chromosome}/{position}/{ref}/{alt}")
+@app.get("/var/{chromosome}-{position}-{ref}-{alt}")
 def get_variant_score(chromosome, position, ref, alt):
-    # TODO call the get_scores function to perform look up, if score not precomputed then call dynamic generation
     scores = get_ditto_score(chrom=chromosome, pos=position, ref=ref, alt=alt)
-    # return {"variant": f"chr{chromosome}:g.{position}{ref}>{alt}"}
-    return {"scores": scores}
+    return {"scores_by_transcript": scores}
 
 
 @app.get("/hgvs/{hgvs_cdna}")

src/utils/predict.py

@@ -5,6 +5,7 @@
 def parse_and_predict(dataframe, config_dict, clf):
     # Drop variant info columns so we can perform one-hot encoding
     dataframe["so"] = dataframe["consequence"].copy()
+    var = dataframe[["transcript","gene","consequence","chrom","pos","ref_base","alt_base"]]
     dataframe = dataframe.drop(config_dict["id_cols"], axis=1)
     dataframe = dataframe.replace([".", "-", ""], np.nan)
     for key in dataframe.columns:
@@ -44,5 +45,9 @@ def parse_and_predict(dataframe, config_dict, clf):
             )
 
     y_score = 1 - clf.predict(df2, verbose=0)
-    del temp_df,df2
-    return y_score
+    y_score = pd.DataFrame(y_score, columns=["DITTO"])
+
+    var = pd.concat([var.reset_index(drop=True), y_score.reset_index(drop=True)], axis=1)
+
+    del temp_df,df2, y_score
+    return var

src/utils/query.py

@@ -81,13 +81,8 @@ def get_ditto_score(chrom: str, pos: int, ref: str, alt: str):
             "Could not get variant annotations from OpenCravat's API. Please check the variant info and try again."
         )
     else:
-        score_dict = {}
-        for transcript in overall["transcript"].unique():
-            transcript_data = overall[overall["transcript"] == transcript].reset_index(
-                drop=True
-            )
-            y_score = parse_and_predict(transcript_data, config_dict, clf)
-            y_score = round(y_score[0][0], 2)
-            score_dict[transcript] = str(y_score)
-        return score_dict
+        var_df_scores = parse_and_predict(overall, config_dict, clf)
+        var_df_scores = var_df_scores.set_index('transcript')
+        var_df_scores = var_df_scores.astype({"DITTO": str, "pos": str})
+        return json.loads(var_df_scores.to_json(orient="index"))