def calculate_region(arg):
gname, sub, prot, start, win, nseqs, trop_dict = arg
treename = 'quicktrees/%s-%s-%s-%i-%i.tree' % (gname, sub, prot, start, win)
matfname = 'quicktrees/%s-%s-%s-%i-%i.pkl' % (gname, sub, prot, start, win)
if os.path.exists(treename):
#benj_res = 'Already Processed'
#return gname, sub, prot, win, start, benj_res
with open(matfname) as handle:
dmat = pickle.load(handle)
with open(treename) as handle:
tree = dendropy.Tree.get_from_stream(handle, 'newick')
else:
is_aa = prot != 'LTR'
alphabet = generic_protein if is_aa else generic_dna
try:
tree, dmat = TreeingTools.phylip_tree_collapse_unique(nseqs, alphabet=alphabet)
except ValueError:
benj_res = 'Too few unique sequences to process'
return gname, sub, prot, win, start, benj_res
except:
benj_res = 'uncaught exception in dist-mat'
return gname, sub, prot, win, start, benj_res
print 'writing'
with open(matfname, 'w') as handle:
pickle.dump(dmat, handle)
with open(treename, 'w') as handle:
tree.write_to_stream(handle, 'newick')
try:
benj_res = TreeingTools.check_distance_pvals(dmat, trop_dict, nreps = 50)
except AssertionError:
benj_res = 'too few groups'
return gname, sub, prot, win, start, benj_res
except:
benj_res = 'uncaught exception'
return gname, sub, prot, win, start, benj_res
try:
out = TreeingTools.evaluate_association_index(tree, trop_dict)
benj_res['AI'], benj_res['AI-pval'], benj_res['AI-null'] = out
except:
benj_res['AI'], benj_res['AI-pval'], benj_res['AI-null'] = ('error', 'error', 'error')
return gname, sub, prot, win, start, benj_res
def rolling_tree_apply(tup):
group_series, seq_series, kwargs = tup
fname = '/home/will/SubCData/Trees/Tree-%(sub)s-%(Prot)s-%(Start)i-%(WinSize)i.newick' % kwargs
if os.path.exists(fname):
return True
alpha = generic_dna if kwargs['Prot'] == 'LTR' else generic_protein
seq_series = seq_series.dropna(thresh = 5)
vseq, vgroup = seq_series.align(group_series.dropna(), join = 'inner', axis = 0)
nseq_ser = vseq.apply(append_seq, axis = 1)
nseqs = sorted(nseq_ser.to_dict().items())
trop_dict = vgroup.to_dict()
#print nseqs
#try:
# tree, dmat = TreeingTools.phylip_tree_collapse_unique(nseqs, alphabet=alpha, use_fast=True)
#except:
# return False
#print 'treeing', fname
tree = TreeingTools.run_FastTree(nseqs, alphabet=alpha, uniq_seqs=True)
with open(fname, 'w') as handle:
tree.write(handle, schema='newick')
return True
try:
tree, dmat = TreeingTools.phylip_tree_collapse_unique(nseqs, alphabet=alpha, use_fast=True)
benj_res = TreeingTools.check_distance_pvals(dmat, trop_dict, nreps = 50)
except:
return kwargs
benj_res.update(kwargs)
try:
out = TreeingTools.evaluate_association_index(tree, trop_dict)
benj_res['AI'], benj_res['AI-pval'], benj_res['AI-null'] = out
except:
benj_res['AI'], benj_res['AI-pval'], benj_res['AI-null'] = (None, None, None)
return benj_res
# <codecell>
# <codecell>
#with open('allgp120.fasta', 'w') as handle:
tres = []
for key, row in wanted_data[['gp120-seq-align', 'Tropism']].dropna().iterrows():
oname = key+'-'+row['Tropism']
tres.append((oname, ''.join(row['gp120-seq-align'])))
# <codecell>
tree, dmat = TreeingTools.phylip_tree_collapse_unique(tres, alphabet=generic_protein)
# <codecell>
with open('gp120tree.nexus', 'w') as handle:
tree.write_to_stream(handle, 'nexus')
# <codecell>
import networkx
with open('gp120tree.dot') as handle:
new_tree = networkx.read_dot(handle)
# <codecell>
pos = networkx.spring_layout(new_tree, dim=100)
