Fix format and annotation issues (#27)

* fixed format issue

* fixed D19 issue

* Fixed D21 error

* fixed D21 typo

* fixed minor error in FGA

* updated strait razor format test data

* fixed partial sequences error

* updated flanking report test file

* Fixed D21 LUS error

* added tests for new formatting rules

* fixed style errors

Co-authored-by: Rebecca Mitchell <[email protected]>

lusSTR/annot.py

@@ -17,6 +17,15 @@
 from lusSTR.repeat import collapse_all_repeats, collapse_repeats_by_length
 from lusSTR.repeat import sequence_to_bracketed_form, split_by_n
 from lusSTR.repeat import reverse_complement, reverse_complement_bracketed
+from pkg_resources import resource_filename
+
+
+def get_str_metadata_file():
+    return resource_filename('lusSTR', 'str_markers.json')
+
+
+with open(get_str_metadata_file(), 'r') as fh:
+    str_marker_data = json.load(fh)
 
 
 def split_sequence_into_two_strings(sequence, repeat_for_split):
@@ -50,15 +59,25 @@ def main(args):
         except IndexError:
             project = 'NA'
             analysis = 'NA'
-
+        metadata = str_marker_data[locus]
+        if (
+            len(sequence) <= (metadata['Foren_5'] + metadata['Foren_3']) and not args.uas
+            and args.kit == 'forenseq'
+        ):
+            flank_summary = [
+                sampleid, project, analysis, locus, reads, 'NA', sequence, 'NA', 'NA', 'NA',
+                'Partial sequence'
+            ]
+            flanks_list.append(flank_summary)
+            continue
         marker = lusSTR.marker.STRMarkerObject(locus, sequence, uas=args.uas, kit=args.kit)
         summary = [sampleid, project, analysis, locus] + marker.summary + [reads]
         list_of_lists.append(summary)
 
         if not args.uas and args.kit == 'forenseq':
             flank_summary = [
                 sampleid, project, analysis, locus, reads, marker.canonical, marker.sequence,
-                marker.flank_5p, marker.annotation, marker.flank_3p
+                marker.flank_5p, marker.annotation, marker.flank_3p, "NA"
             ]
             flanks_list.append(flank_summary)
 
@@ -71,8 +90,9 @@ def main(args):
     name = os.path.splitext(args.out)[0]
     if not args.uas:
         flanks_columns = [
-            'SampleID', 'Project', 'Analysis', 'Locus', 'Reads', 'Length_Allele', 'Full_Sequence',
-            '5_Flank_Anno', 'UAS_Region_Anno', '3_Flank_Anno'
+            'SampleID', 'Project', 'Analysis', 'Locus', 'Reads', 'Length_Allele',
+            'Full_Sequence', '5_Flank_Anno', 'UAS_Region_Anno', '3_Flank_Anno',
+            'Potential_Issues'
         ]
         final_flank_output = pd.DataFrame(flanks_list, columns=flanks_columns)
         final_flank_output.to_csv(f'{name}_flanks_anno.txt', sep='\t', index=False)

lusSTR/format.py

@@ -69,7 +69,7 @@ def main(args):
         path = args.input
         analysisID = path.rstrip(os.sep)
         analysisID_final = os.path.basename(analysisID)
-        results_final['Project'] = "NA"
+        results_final['Project'] = analysisID_final
         results_final['Analysis'] = analysisID_final
 
     output_file = sys.stdout

lusSTR/marker.py

@@ -9,18 +9,13 @@
 
 import json
 import lusSTR
-from lusSTR.annot import split_sequence_into_two_strings
+from lusSTR.annot import get_str_metadata_file, split_sequence_into_two_strings
 from lusSTR.repeat import collapse_repeats_by_length, sequence_to_bracketed_form
 from lusSTR.repeat import reverse_complement, reverse_complement_bracketed
 from lusSTR.repeat import repeat_copy_number, collapse_all_repeats, split_by_n
-from pkg_resources import resource_filename
 import re
 
 
-def get_str_metadata_file():
-    return resource_filename('lusSTR', 'str_markers.json')
-
-
 with open(get_str_metadata_file(), 'r') as fh:
     str_marker_data = json.load(fh)
 
@@ -146,6 +141,15 @@ def canonical(self):
             canon_allele = f'{allele_int}.{allele_dec}'
         return canon_allele
 
+    @property
+    def indel_flag(self):
+        '''Check for potential indels within flanking regions'''
+        if str(self.canonical) not in self.data['Alleles']:
+            flag = 'Possible indel or partial sequence'
+        else:
+            flag = ' '
+        return flag
+
     @property
     def cannot_split(self):
         return self.locus in [
@@ -454,7 +458,7 @@ def annotation(self):
         annotation which is consistent with previously published annotation for this locus.
         '''
         sequence = self.forward_sequence
-        if len(sequence) % self.repeat_size == 0:
+        if len(sequence) % self.repeat_size == 0 or (not ('GGAA') in sequence):
             return collapse_repeats_by_length(sequence, self.repeat_size)
         else:
             final = list()
@@ -565,7 +569,9 @@ def annotation(self):
             prev = m.end()
         if (
             prev == (len(forward_strand_brack_form) - 1) or
-            prev == (len(forward_strand_brack_form) - 2)
+            prev == (len(forward_strand_brack_form) - 2) or
+            prev == (len(forward_strand_brack_form) - 4) or
+            prev == (len(forward_strand_brack_form) - 5)
            ):
             return forward_strand_brack_form
         else:
@@ -619,8 +625,12 @@ def designation(self):
                         repeats = repeat_copy_number(i, repeat)
                         lus_sec.append(repeats)
         if lus_allele is None:
-            lus_allele = lus_sec[1]
-            sec_allele = lus_sec[0]
+            if len(lus_sec) == 2:
+                lus_allele = lus_sec[1]
+                sec_allele = lus_sec[0]
+            else:
+                lus_allele = 0
+                sec_allele = lus_sec[0]
 
         finalcount = 0
         for m in re.finditer(self.data['Tert'], self.annotation):
@@ -690,13 +700,15 @@ def annotation(self):
             final.append(sequence_to_bracketed_form(first_string, 4, self.repeats))
         else:
             final.append(collapse_repeats_by_length(first_string, 4))
-        if (len(second_string) % 4 != 0):
-            third_string = second_string[:-6]
-            final.append(collapse_repeats_by_length(third_string, 4))
-            final.append(second_string[-6:-4])
-            final.append(second_string[-4:])
-        else:
-            final.append(collapse_repeats_by_length(second_string, 4))
+        if (second_string != ""):
+            if (len(second_string) % 4 != 0):
+                if (len(second_string) > 6):
+                    third_string = second_string[:-6]
+                    final.append(collapse_repeats_by_length(third_string, 4))
+                final.append(second_string[-6:-4])
+                final.append(second_string[-4:])
+            else:
+                final.append(collapse_repeats_by_length(second_string, 4))
         final_string = ' '.join(final)
         return re.sub(r' +', ' ', final_string)