def ali_in_tree(self,aliname = 'group2.stk',
rank = 'genus',
**kwargs):
all_seqs = ali.get_seqs(aliname)
alinodes = ali.get_taxnodes(aliname)
aliranks = [t.rank if t else None for t in alinodes]
all_leaves = self.t.get_terminals()
leafnodes = self.leafNodes(reset = mod(reset, 2))
leafranks =[n.rank if n else None for n in leafnodes]
ali_families = ali.get_taxon_forall(rank = rank,aliname = aliname,
**mem.sr(kwargs))
leaf_families= self.getTaxon(rank = rank, **mem.sr(kwargs))
aset = set(ali_families)
lset = set(leaf_families)
a_domains =[(node, ncbi.get_taxon(node,'superkingdom')) for node in aset]
l_domains =[(node, ncbi.get_taxon(node,'superkingdom')) for node in lset]
bac_domain = [x[1] for x in l_domains if ncbi.sciname(x[1])== 'Bacteria'][0]
l_bacs = set((l[0] for l in l_domains if l[1] == bac_domain))
a_bacs = set((a[0] for a in a_domains if a[1] == bac_domain))
leaf_bacteria = [leaf if leaf in l_bacs else None for leaf in leaf_families]
ali_bacteria = [a if a in a_bacs else None for a in ali_families]
return leaf_bacteria, ali_bacteria, leafnodes, alinodes
def show_rank(self,
tree_n_rank, ali_n_rank,
tree_n, ali_n
):
tree_leaves =self.t.get_terminals()
rank_unq = set(tree_n_rank).union(set(ali_n_rank))
supertaxa = {}
for u in rank_unq:
if not u: continue
tree_matches = nonzero(equal(tree_n_rank, u))[0]
tree_leaf_subset = [tree_leaves[i] for i in tree_matches]
tree_n_subset = [tree_n[i] for i in tree_matches]
ali_matches = nonzero(equal(ali_n_rank, u))[0]
ali_n_subset = [ali_n[i] for i in ali_matches]
supertaxa[ncbi.sciname(u)] = (tree_n_subset, ali_n_subset)
f = plt.figure(0)
f.clear()
xax = arange(len(supertaxa))
ax = f.add_subplot('111',xticks = [])
plots = [ax.plot(xax,[log(len(e[0])+1) for e in supertaxa.values()],
linewidth = 2, label='tree')[0],
ax.plot(xax,[log(len(e[1])+1) for e in supertaxa.values()],
linewidth = 2, label='ali')[0]]
ax.legend(plots, [x.get_label() for x in plots])
for i,x in enumerate(xax):
ax.text(x,0,supertaxa.keys()[i],
va = 'top', ha = 'left',
rotation = -15.)
def __repr__(self):
return '{0},{1},{2} ({3})'.format(ncbi.sciname(self.family_node),
ncbi.sciname(self.species_node),
ncbi.sciname(self.term_node),
self.term_node.rank)
def node_repr(self):
import compbio.projects.cbdb.helpers.ncbi as ncbi
sn = ncbi.sciname(self)
if sn: return '{0:15} ({1}) '.format(sn, self.rank)
elif self.names: return '{0:15} ({1})'.format(sn, self.rank)
else: return '{0:30} ({1:15})'.format('[unnamed ncbi]',self.rank)