NimaGen 29-plex (#134)

.gitattributes

@@ -1,2 +1,4 @@
 microhapdb/_version.py export-subst
 notebooks/*.ipynb linguist-documentation
+dbbuild/sources/auton2015/scripts/infocalc linguist-vendored
+dbbuild/legacy/sources/1kgp/infocalc linguist-vendored

dbbuild/indels.csv

@@ -1,7 +1,7 @@
 Marker,VariantIndex,Refr,Alt
 mh01ZBF-002,0,T,TT
-mh02FHL-003.v3,3,A,AC;ACC
-mh02FHL-006.v2,33,T,TA
+mh02FHL-003,3,A,AC;ACC
+mh02FHL-006,33,T,TA
 mh02ZBF-001,2,C,CC
 mh03FHL-003.v1,1,AT,A
 mh03ZBF-001,0,CAATCAATCAA,CAATCAA

dbbuild/lib/interval.py

@@ -29,7 +29,7 @@ def check(self):
                 loci = set([m.locus for m in interval.data])
                 if len(loci) == 1:
                     continue
-                markers = sorted(interval.data, key=lambda m: m.sortkey)
+                markers = sorted(interval.data, key=lambda m: m.sources[0].sortkey)
                 for marker in markers[1:]:
                     if marker.name != markers[0].name and marker.name not in self.mergeables:
                         self.mergeables[marker.name] = markers[0].name

dbbuild/lib/marker.py

@@ -11,6 +11,7 @@
 # -------------------------------------------------------------------------------------------------
 
 from .variant import VariantList
+from itertools import chain
 import pandas as pd
 
 
@@ -178,6 +179,15 @@ def overlaps(self, other):
         same_chrom = self.chrom_num == other.chrom_num
         return same_chrom and self.start <= other.end and self.end >= other.start
 
+    def rsid_union(self, *others):
+        rsids = set(self.rsids)
+        for other in others:
+            rsids |= set(other.rsids)
+        for rsid in rsids:
+            for marker in chain([self], others):
+                if rsid not in marker.rsids:
+                    marker.rsids.append(rsid)
+
     @property
     def sortkey(self):
         return self.chrom_num, self.span, self.name