import os,re,sys,miscMySQL,miscTaxonomy
from collections import defaultdict
from decimal import Context, localcontext
conn = miscMySQL.get_conn_ncRNA()
d_cur = miscMySQL.get_dict_cursor(conn)
connNCBI = miscMySQL.get_conn_NCBI()
curNCBI = connNCBI.cursor()
def tally():
result = defaultdict(lambda: defaultdict(lambda: ''))
d_cur.execute("select family,count(*) as total from Zasha_20081002_plus20071102_curated group by family")
for r in d_cur.fetchall():
result[r['family']]['total'] = r['total']
d_cur.execute("select z.family as family,c.phylum as phylum, count(*) as count from Zasha_20081002_plus20071102_curated as z left join NCBI.cache_acc_to_tax as c on (z.acc=c.acc) group by z.family,c.phylum")
for r in d_cur.fetchall():
result[r['family']][r['phylum']] = r['count']
d_cur.execute("select distinct phylum from NCBI.cache_acc_to_tax order by phylum")
phyla = map(lambda x: x['phylum'],d_cur.fetchall())
print("\tTOTAL\t"+"\t".join(phyla))
for family,d in result.iteritems():
print("{0}\t{1}".format(family,d['total'])),
for p in phyla:
print("\t"+str(d[p])),
print ''
tally()
sys.exit(-1)
# here's what we're interested in:
from miscMySQL import get_conn_Actino, get_dict_cursor
from collections import defaultdict
from operator import itemgetter
from scipy import histogram
"""
This tmp script is used to do evaluation on the cliques we generated
"""
clique_filename='output/output_cliques/ALLActino_RefSeq25_m30s0_cut35gamma80.cliques.pickle'
# here we keep track of some stats
ncRNA_ids_seen = defaultdict(lambda: 0)
clique_stats = defaultdict(lambda: {'sizes':[],'precisions':[]})
ncRNA_id_to_family = {}
hitQ_sizes = []
conn = get_conn_Actino()
cursor = get_dict_cursor(conn)
with open(clique_filename) as f:
QQQ = load(f)
for Q in QQQ:
if 75 in Q or 163 in Q: continue
if len(Q) < 5:
continue
tally_by_family = defaultdict(lambda: 0)
for i in Q:
# if it's not a hit, will return (None,None)
(ncRNA_id, ncRNA_family) = c2.check_hit(i, cursor)
tally_by_family[ncRNA_family] += 1
ncRNA_ids_seen[ncRNA_id] += 1
ncRNA_id_to_family[ncRNA_id] = ncRNA_family
# decide the dominant family of this cluster
tally_by_family = tally_by_family.items()