def assign_amino_acid_muts_vcf(prots, path):
tree_meta = read_tree_meta_data(path)
seqNames = prots[prots.keys()[0]]['sequences'].keys()
excluded = []
#go through every gene in the prots nested dict
for fname, prot in prots.iteritems():
sequences = prot['sequences']
ref = prot['reference']
positions = prot['positions']
pats = []
i=0
#for each position, get the mutation in the right format
#[ancestral][position][mutation]
while i < len(positions):
pi = positions[i]
pos = pi+1 #convert to standard numbering for output (# starts at 1)
refb = ref[pi]
pattern = [ refb+str(pos)+sequences[k][pi] if pi in sequences[k].keys()
else "" for k,v in sequences.iteritems() ]
#if the exact same mutation in all sequences, don't include! (only mutant against ref..)
if not (len(pattern)==len(sequences) and len(np.unique(pattern))==1):
pats.append(pattern)
i+=1
#convert our list of lists to matrix
patMat = np.matrix(pats)
#don't include if all the mutations identical across sequences! (only mutant against ref..)
if len(pats) != 0:
#for every sequence, assign the mutations in tree_meta
for i in xrange(len(seqNames)):
node_name = seqNames[i]
ary = np.array(patMat[:,i]).reshape(-1,)
tree_meta[node_name][fname+'_mutations'] = ",".join(ary[ary != ''])
else:
excluded.append(fname)
if len(excluded) != 0:
print "{} genes do not differ across the tree. They will not be added to tree meta-data or shown in auspice".format(len(excluded))
#write it out!
write_tree_meta_data(path, tree_meta)
def add_drm_tree_meta(path, seqDRM):
tree_meta = read_tree_meta_data(path)
#add drug resistance to tree_meta, & make list for colouring
drugMuts = {}
drugMuts["Drug_Resistance"] = ['0']
for seq, v in seqDRM.iteritems():
#in case re-running, don't add mutations to old ones!
remove_old_DRM(tree_meta[seq])
tempList = {}
for mut, drug in v.iteritems():
drugs = drug.split(';')
for drug in drugs:
trDrug = drugTranslate(drug)
if trDrug in tree_meta[seq]:
tree_meta[seq][trDrug] = ",".join(
[tree_meta[seq][trDrug], mut])
else:
tree_meta[seq][trDrug] = mut
if trDrug in drugMuts:
if tree_meta[seq][trDrug] not in drugMuts[trDrug]:
drugMuts[trDrug].append(tree_meta[seq][trDrug])
else:
drugMuts[trDrug] = [tree_meta[seq][trDrug]]
tempList[trDrug] = ""
numResist = str(len(tempList))
tree_meta[seq]["Drug_Resistance"] = numResist
if numResist not in drugMuts["Drug_Resistance"]:
drugMuts["Drug_Resistance"].append(numResist)
#for any with no resistance, add a 0 to tree_meta
for seq, v in tree_meta.iteritems():
if 'Drug_Resistance' not in tree_meta[seq]:
tree_meta[seq]["Drug_Resistance"] = '0'
write_tree_meta_data(path, tree_meta)
return drugMuts
#T = tree_newick(path)
meta = read_sequence_meta_data(path)
fields = ['branchlength', 'clade']
if args.timetree:
tt = timetree(tree=T, aln=ref_alignment(path), confidence=args.confidence,
seq_meta=meta, reroot=None if args.keeproot else 'best',Tc=args.Tc)
T = tt.tree
fields.extend(['mutations', 'mutation_length', 'num_date', 'clock_length'])
if args.confidence:
fields.append('num_date_confidence')
elif args.ancestral:
tt = ancestral_sequence_inference(tree=T, aln=ref_alignment(path))
T = tt.tree
fields.extend(['mutations', 'mutation_length'])
clade_index = 0
for n in T.find_clades(order='preorder'):
n.clade = clade_index
clade_index+=1
Phylo.write(T, tree_newick(path), 'newick')
meta_dic = collect_tree_meta_data(T, fields)
write_tree_meta_data(path, meta_dic)
with open(sequence_gtr_model(path),'w') as ofile:
ofile.write(str(tt.gtr))
if args.timetree or args.ancestral:
export_sequence_fasta(T, path)
seqs[seq.name] = seq
muts = {}
muts[T.root.name]=''
for node in T.get_nonterminals():
pseq = seqs[node.name]
for c in node:
cseq = seqs[c.name]
muts[c.name]=','.join([anc+str(pos+1)+der
for pos, (anc, der) in enumerate(zip(pseq, cseq))
if anc!=der])
return muts
if __name__ == '__main__':
parser = generic_argparse("Assign amino acid mutations to the tree")
args = parser.parse_args()
path = args.path
tree_meta = read_tree_meta_data(path)
T = Phylo.read(tree_newick(path), 'newick')
for gene, aln_fname in get_genes_and_alignments(path, tree=True):
if gene!='nuc':
muts = get_amino_acid_mutations(T, aln_fname)
for node_name in tree_meta:
tree_meta[node_name][gene+'_mutations'] = muts[node_name]
write_tree_meta_data(path, tree_meta)
