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 make_anc_batches(aliname, rank_name = 'phylum',
do_bsub = False,
run = False, nrun = 0,
**kwargs):
BT = getBTOL(**mem.sr(kwargs))
tree_tax = BT.getTaxon(rank_name)
ali_tax = ali.get_taxon_forall(aliname, rank_name)
union_tax = set(tree_tax).intersection(set(ali_tax))
union_taxids = [e.id for e in union_tax if e]
batch_pdicts = [dict(taxid = taxid,
rank_name = rank_name,
aliname = aliname)
for taxid in union_taxids]
cmds = []
for idx, d in enumerate(batch_pdicts):
run_id=bsub.get_run_id(idx, prefix = rank_name)
bsub.save_inp(d, run_id)
cmds.append(bsub.cmd(os.path.abspath(inspect.stack()[0][1]),'run_anc',run_id,\
do_bsub = do_bsub, run_id = run_id))
if run:
for c in cmds:
out = subprocess.call(c, shell = True)
print out
return cmds
def set_taxnodes(**kwargs):
all_seqs = get_seqs(dbname,**mem.sr(kwargs))
seq_taxa = [s.source_taxon
if s.source_taxon else None
for s in all_seqs]
alinodes = [ncbi.get_node(s) if s != None else None for s in seq_taxa]
return alinodes
def setBTOL(**kwargs):
B = BTOL(**mem.sr(kwargs))
if not B.treeInitialized():
print 'Underlying tree structure apparently uninitialized: initializing\n...'
B.initTree()
print '...\nDone\nSaving\n...'
B.saveTree()
print '...\nDone'
return B
def check_bdtnp(**kwargs):
ids = id_map(**mem.sr(kwargs))
btgs = nio.getBDTNP()
stfs,stgs = nio.getNet()
sxs = dict([(i, {}) for i in ids.keys()])
for i, elt in enumerate(ids.iteritems()):
gid, gene_val = elt
if stfs.has_key(gid) : sxs[gid]['stf'] = True;
if btgs.has_key(gid) : sxs[gid]['btg'] = True;
if stgs.has_key(gid) : sxs[gid]['stg'] = True;
return sxs
def getTaxon(self,rank = rank,
**kwargs):
def setTaxon(BTInstance = None, rank = None, **kwargs):
assert rank; assert BTInstance
leafnodes = BTInstance.leafNodes(**mem.sr(kwargs))
leaf_families = [ncbi.get_taxon(node, rank=rank)
if node else None for node in leafnodes]
return leaf_families
return mem.getOrSet(setTaxon,
**mem.sr(kwargs,
rank = rank,
BTInstance = self,
on_fail = 'compute',
hardcopy = False,
register = rank))
def run(rfid,run_id, inp_run_id, reset = True,
draw_alis = draw_all_easy):
sgs = get_seq_groups(rfid = rfid, **mem.sr({},reset = False))
all_seq_group_datas = []
for s in sgs:
if len(s) < 4:
all_seq_group_datas.append(None)
continue
if len(s) > 40:
print 'skipping cuz it takes too long!'
continue
print 'Not Skipping'
all_seq_group_datas.append(eval_seq_group(s,rfid, '{0}_{1}'.format(run_id,len(s)), inp_run_id, reset = reset,
draw_alis = draw_alis))
return all_seq_group_datas
def run(**kwargs):
BT = getBTOL(**mem.sr(kwargs))
seqnodes = BT.investigatePhylum(**kwargs)
recs, seqelts, seqtuples = seq_recs(seqnodes)
align = align_seqnodes(recs)
tree = phyml.tree(align)
rstfile= paml.run_paml(tree, align)
anc_tree = paml.rst_parser(rstfile)
anc_alignment = [SeqRecord(elt.m['seq'],
id = None,
name = elt.name,
annotations = {'scores':elt.m['probs']})
for elt in anc_tree.get_nonterminals()]
return (tree, anc_tree), (align, anc_alignment)
def set_datafiles(**kwargs):
out ={}
idmap = id_map(**mem.sr(kwargs))
for k,v in idmap.iteritems():
out[k] = array([ [float(e) for e in re.compile('\s+').split(l.strip())] for l in open(v['file']).readlines() if l[0] in '0123456789'])
return out
def __init__(self,
**kwargs):
self.t = sqt.init(**mem.sr(kwargs))
def investigatePhylum(self,
aliname = 'group2.stk',
p_node = None, **kwargs):
if not p_node: p_node = ncbi.taxon_with_name('phylum', 'Thermotogae')
ali_seqs = ali.get_seqs(aliname, **mem.sr(kwargs))
ali_nodes = array(ali.get_taxnodes(aliname, **mem.sr(kwargs)))
ali_phyla = array(ali.get_taxon_forall(aliname,**mem.sr(kwargs, rank = 'phylum')))
ali_inds = nonzero(equal(ali_phyla, p_node))[0]
leaf_terminals = self.t.get_terminals()
leaf_nodes = array(self.leafNodes(**mem.sr(kwargs)))
leaf_phyla = array(self.getTaxon('phylum', **mem.sr(kwargs)))
leaf_inds = nonzero(equal(leaf_phyla, p_node))[0]
ap_sub = ali_phyla[ali_inds]
lp_sub = leaf_phyla[leaf_inds]
ag_sub = array(ali.get_taxon_forsome(ali_nodes[ali_inds],'genus','thermo',
**mem.sr(kwargs)))
lg_sub = array(self.getTaxon('genus', **mem.sr(kwargs)))[leaf_inds]
as_sub = array(ali.get_taxon_forsome(ali_nodes[ali_inds], 'species', 'thermo'))
ls_sub = array(self.getTaxon('species',**mem.sr(kwargs)))[leaf_inds]
db16 = cbdb.getName('16s')
a_16s= [ db16.S.q(db16.Sequence).
filter_by(source_taxon = n.id).all() for n in ali_nodes[ali_inds]]
l_16s= [ db16.S.q(db16.Sequence).
filter_by(source_taxon = n.id).all() for n in leaf_nodes[leaf_inds]]
#fill any empty nodes... (those lacking 16s rRNA)
for idx, elt in enumerate(a_16s):
cur_node= ali_nodes[ali_inds[idx]]
while not elt:
cur_node = cur_node.parent
elt.extend(db16.S.q(db16.Sequence).filter_by(source_taxon = cur_node.id).all())
for idx, elt in enumerate(l_16s):
cur_node= leaf_nodes[leaf_inds[idx]]
while not elt:
cur_node = cur_node.parent
elt.extend(db16.S.q(db16.Sequence).filter_by(source_taxon = cur_node.id).all())
all_lens = dict([ (k, [len(list( e)) for e in seqlist] )
for seqlist,k in [[a_16s,'a_16s'],[l_16s,'l_16s']]])
leaf_sns = [ SeqNode(lg_sub[i],ls_sub[i] , leaf_nodes[idx],
[(x.sequence,x.gb_id, x.source_taxon, ncbi.get_node(x.source_taxon).rank)
for x in l_16s[i]],
src = leaf_terminals[idx],
node_id = 'btol:default:{0}'.format(leaf_terminals[idx].m['id']))
for i, idx in enumerate(leaf_inds)]
ali_sns = [ SeqNode(ag_sub[i],as_sub[i] , ali_nodes[idx],
[( x.sequence,x.gb_id , x.source_taxon, ncbi.get_node(x.source_taxon).rank)
for x in a_16s[i]],
src = ali_seqs[idx],
node_id = 'ali:{0}:{1}'.format(aliname,ali_seqs[idx].id))
for i, idx in enumerate(ali_inds)]
return list(it.chain(leaf_sns, ali_sns))
def setTaxon(BTInstance = None, rank = None, **kwargs):
assert rank; assert BTInstance
leafnodes = BTInstance.leafNodes(**mem.sr(kwargs))
leaf_families = [ncbi.get_taxon(node, rank=rank)
if node else None for node in leafnodes]
return leaf_families
def setTaxon(aliname = None, rank = None,**kwargs):
assert aliname != None and rank != None
nodes = get_taxnodes(aliname,**mem.sr(kwargs))
taxon = [ncbi.get_taxon(node, rank=rank)
if node else None for node in nodes]
return taxon
def setC2(**kwargs): ll = c2(**mem.sr(kwargs)) result = c2(ll, **mem.sr(kwargs)) return result