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
if __name__ == '__main__':
parser = generic_argparse("Export precomputed data as auspice jsons")
parser.add_argument(
'--prefix',
required=True,
help=
"prefix for json files that are passed on to auspice (e.g., zika.fasta)"
)
parser.add_argument(
'--reference',
required=True,
help="reference sequence needed for entropy feature export")
args = parser.parse_args()
path = args.path
T = Phylo.read(tree_newick(path), 'newick')
seq_meta = read_sequence_meta_data(path)
tree_meta = read_tree_meta_data(path)
attach_tree_meta_data(T, tree_meta)
tree_layout(T)
fields_to_export = tree_meta.values()[0].keys() + [
"tvalue", "yvalue", "xvalue", "attr", "muts", "aa_muts"
]
tjson = tree_to_json(T.root, extra_attr=fields_to_export)
write_json(tjson, tree_json(path, args.prefix))
export_sequence_json(T, path, args.prefix)
export_diversity(path, args.prefix, args.reference)