add cli options to assign pangenome outgroups

Author: glennhickey

src/cactus/refmap/cactus_pangenome.py

@@ -52,7 +52,8 @@ def main():
     parser.add_argument("seqFile", help = "Seq file (will be modified if necessary to include graph Fasta sequence)")
     parser.add_argument("--outDir", help = "Output directory", required=True)
     parser.add_argument("--outName", help = "Output name (without extension)", required=True)
-    parser.add_argument("--reference", required=True, nargs='+', type=str, help = "Reference event name(s). The first will be the \"true\" reference and will be left unclipped and uncollapsed. It also should have been used with --reference in all upstream commands. Other names will be promoted to reference paths in vg") 
+    parser.add_argument("--reference", required=True, nargs='+', type=str, help = "Reference event name(s). The first will be the \"true\" reference and will be left unclipped and uncollapsed. It also should have been used with --reference in all upstream commands. Other names will be promoted to reference paths in vg")
+    parser.add_argument("--outgroup", type=str, nargs='+', help = "Use given genome as outgroup. Multiple allowed")
 
     # cactus-minigraph options
     parser.add_argument("--mgCores", type=int, help = "Number of cores for minigraph construction (defaults to the same as --maxCores).")

src/cactus/setup/cactus_align.py

@@ -69,6 +69,7 @@ def main():
     parser.add_argument("--outGFA", action="store_true", help = "export pangenome grpah in GFA (.gfa.gz) in addition to HAL")
     parser.add_argument("--batch", action="store_true", help = "Launch batch of alignments.  Input seqfile is expected to be chromfile as generated by cactus-graphmap-slit. IMPORTANT: paffile argument should bot be specified when using this option")    
     parser.add_argument("--reference", type=str, help = "Ensure that given genome is acyclic by deleting all paralogy edges in postprocessing, also do not mask its PAF mappings")
+    parser.add_argument("--outgroup", type=str, nargs='+', help = "Use given genome as outgroup. Only works with --pangenome. Multiple allowed")
 
     #Progressive Cactus Options
     parser.add_argument("--configFile", dest="configFile",
@@ -130,7 +131,10 @@ def main():
             raise RuntimeError('--consCores required for non single_machine batch systems')
     if options.maxCores is not None and options.consCores > int(options.maxCores):
         raise RuntimeError('--consCores must be <= --maxCores')
-    
+
+    if options.outgroup and not options.pangenome:
+        raise RuntimeError('--outgroup can only be used with --pangenome')
+
     options.buildHal = True
     options.buildFasta = True
 
@@ -261,6 +265,9 @@ def make_align_job(options, toil, config_wrapper=None, chrom_name=None):
     if options.reference and options.pangenome:
         # validate the sample names
         check_sample_names(input_seq_map.keys(), options.reference)
+
+    if options.outgroup and options.pangenome:
+        og_map[ options.root if options.root else mc_tree.getRootName()] = options.outgroup
 
     # apply path overrides.  this was necessary for wdl which doesn't take kindly to
     # text files of local paths (ie seqfile).  one way to fix would be to add support