Add new validation case

validate.py

@@ -41,18 +41,7 @@ class Validation():
                     break
             i += 1
 
-        # Check if there are multiple rsIDs
-        i = 0
-        while i < len(df):
-            rsid = df.loc[i, 'rsID']
-            if ',' in rsid:
-                rsids = rsid.split(',')
-                df.loc[i + 0.1], df.loc[i + 0.9] = df.loc[i], df.loc[i]
-                df = df.sort_index().reset_index(drop=True)
-                df.loc[i + 1, 'rsID'], df.loc[i + 2, 'rsID'] = rsids[0], s.join(rsids[1:])
-            i += 1
-
-        # Check if there is SNPs without 'rs'
+        # Check if there are SNPs not well captured
         for i in df.index:
             safe = True
             snp = df.loc[i, 'rsID']
@@ -65,6 +54,10 @@ class Validation():
                 snp = 'r' + snp
             elif re.fullmatch('(\d)+', snp):
                 snp = 'rs' + snp
+            elif re.fullmatch('r(\d)+', snp):
+                snp = 'rs' + snp[1:]
+                if snp[2] == '5':
+                    snp += f',rs{snp[3:]}'
             else:
                 safe = False
                 df = df.drop(i)
@@ -72,6 +65,17 @@ class Validation():
             if safe:
                 df.loc[i, 'rsID'] = snp
 
+        # Check if there are multiple rsIDs
+        i = 0
+        while i < len(df):
+            rsid = df.loc[i, 'rsID']
+            if ',' in rsid:
+                rsids = rsid.split(',')
+                df.loc[i + 0.1], df.loc[i + 0.9] = df.loc[i], df.loc[i]
+                df = df.sort_index().reset_index(drop=True)
+                df.loc[i + 1, 'rsID'], df.loc[i + 2, 'rsID'] = rsids[0], s.join(rsids[1:])
+            i += 1
+
         df.reset_index(drop=True, inplace=True)
         df_clean = df.copy()