bug for single mutation. reduce warning of refseq

fix bug for chromosome with only one mutations

Author: ATPs

src/perChrom.py

@@ -348,6 +348,7 @@ def get_df_transcript2(file_gtf, file_protein, file_genome, cpu_counts, datatype
             print('chromosome', chromosome, list(tdf_special.index), 'not translated from the CDS sequences in the genome. do not change')
             for transcript_id in tdf_special.index:
                 df_transcript2.at[transcript_id, 'mutations'] = []
+            df_transcript2 = df_transcript2[~df_transcript2.index.isin(set(tdf_special.index))]
 
     # split df_transcript2 to with mutations and no_mutations
     df_transcript_noMut = df_transcript2[df_transcript2['mutations'].apply(lambda x:len(x)==0)]

src/perChromSqlite.py

@@ -122,6 +122,8 @@ def run_perChrom(self, save_results = True):
 
         # get df_transcript2 based on dc_transcript2mutations
         protein_ids = tuple(dc_transcript2mutations.keys())
+        if len(protein_ids) == 1:
+            protein_ids = '("{}")'.format(protein_ids[0])
         query = f"SELECT * FROM protein_description WHERE protein_id IN {protein_ids}"
         df_transcript2 = pd.read_sql_query(query, self.con)        # assign mutations to each transcript. 
         if df_transcript2.shape[0] == 0:
@@ -145,6 +147,7 @@ def run_perChrom(self, save_results = True):
                 print('chromosome', chromosome, list(tdf_special.index), 'not translated from the CDS sequences in the genome. do not change')
                 for transcript_id in tdf_special.index:
                     df_transcript2.at[transcript_id, 'mutations'] = []
+                df_transcript2 = df_transcript2[~df_transcript2.index.isin(set(tdf_special.index))]
 
         df_transcript3 = df_transcript2