Merge branch 'dev'

to get to v0.0.2 due to big SUB recruitment bug

README.md

@@ -1,2 +1,27 @@
-# 0039_mOTUlizer
+# mOTUlizer
 Utility to analyse a group of closely related MAgs/Genomes/bins/SUBs of more or less dubious origin
+
+## INSTALL
+
+```
+pip install mOTUlizer
+```
+
+
+## USAGE
+
+### EASY
+
+To make OTUs and get some stats, needs fastANI in the `PATH`, and output of checkm
+
+```
+mOTUlize.py -k checkm_output.txt  --output a_messy_json-file_with_the_output.json
+```
+
+Loads of little options if you do : `mOTUlize.py -h`
+
+Also there is `mOTUpan.py` that can compute core genomes and pangenomes. Needs to be more debugged so try  out at own risk:
+
+```
+mOTUpan.py -h
+```

mOTUlizer/bin/mOTUlize.py

@@ -93,7 +93,7 @@ def main(args):
 
 
 
-    assert all([g in checkm_info  for g in genomes]), "you do not have completness/contamination info for all you bins"
+    assert all([g in checkm_info  for g in genomes]), "you do not have completness/contamination info for all you bins, values missing for :" + ", ".join([g for g in genomes if g not in checkm_info][0:10]) + "... (only 10 first shown   )"
 
     if fnas == {}:
         fnas = {g : None for g in genomes}
@@ -137,7 +137,7 @@ def main(args):
     parser.add_argument('--MAG-completeness', '--MC', '-M', nargs = '?', type=float, default = 40, help = "completeness cutoff for seed MAGs, default : 40")
     parser.add_argument('--MAG-contamination', '--Mc', '-m', nargs = '?', type=float, default = 5, help = "contamination cutoff for seed MAGs, default : 5")
     parser.add_argument('--SUB-completeness', '--SC', '-S', nargs = '?', type=float, default = 0, help = "completeness cutoff for recruited SUBs, default : 0")
-    parser.add_argument('--SUB-contamination', '--Sc', '-s', nargs = '?', type=float, default = 0, help = "contamination cutoff for recruited SUBs, default : 0")
+    parser.add_argument('--SUB-contamination', '--Sc', '-s', nargs = '?', type=float, default = 100, help = "contamination cutoff for recruited SUBs, default : 0")
     parser.add_argument('--similarity-cutoff', '-i', nargs = '?', type=float, default = 95, help = "distance cutoff for making the graph, default : 95")
     parser.add_argument('--cpus', '-c', nargs = '?', type=int, default = 1, help = "number of threads, default : 1")
     parser.add_argument('--keep-simi-file', '-K', nargs = '?', default = None, help = "keep generated similarity file if '--similarities' is not procided")

mOTUlizer/classes/mOTU.py

@@ -245,6 +245,7 @@ def __from_bins(self, bins, name,dist_dict = None ):
         self.core = None
         self.fastani_dict = dist_dict
         self.aa2cog = None
+        self.likelies = None
 
 
     @classmethod
@@ -258,7 +259,7 @@ def cluster_MetaBins(cls , all_bins, dist_dict, ani_cutoff = 95, prefix = "mOTU_
         tt = [(k, v.checkm_complet, v.checkm_contamin) for k, v in all_bins.items() if v.checkm_complet > 0]
 
         good_mag = lambda b : all_bins[b].checkm_complet > mag_complete and all_bins[b].checkm_contamin < mag_contamin
-        decent_sub = lambda b : all_bins[b].checkm_complet > sub_complete and all_bins[b].checkm_contamin < mag_contamin and not good_mag(b)
+        decent_sub = lambda b : all_bins[b].checkm_complet > sub_complete and all_bins[b].checkm_contamin < sub_contamin and not good_mag(b)
         good_pairs = [k for k,v  in dist_dict.items() if v > ani_cutoff and dist_dict.get((k[1],k[0]), 0) > ani_cutoff and good_mag(k[0]) and good_mag(k[1])]
         species_graph = igraph.Graph()
         vertexDeict = { v : i for i,v in enumerate(set([x for k in good_pairs for x in k]))}
@@ -277,12 +278,16 @@ def cluster_MetaBins(cls , all_bins, dist_dict, ani_cutoff = 95, prefix = "mOTU_
 
         left_pairs = {k : v for k, v in dist_dict.items() if v > ani_cutoff and k[0] != k[1] and ((decent_sub(k[0]) and good_mag(k[1])) or (decent_sub(k[1]) and good_mag(k[0])))}
         print("looking for good_left pairs")
-        subs = {l[0] : (None,0) for l in left_pairs}
+#        print(left_pairs)
 
+        subs = {l : (None,0) for ll in left_pairs.keys() for l in ll if not good_mag(l)}
+#        print(subs)
         print("looking for best mOTU match")
         for p,ani in left_pairs.items():
-            if subs[p[0]][1] < ani:
+            if p[0] in subs and subs[p[0]][1] < ani:
                 subs[p[0]] = (p[1], ani)
+            if p[1] in subs and subs[p[1]][1] < ani:
+                subs[p[1]] = (p[0], ani)
 
         genome_clusters = [set(gg) for gg in genome_clusters]