def check_login(form_data):
db = DB()
user = unicode(form_data['user'])
result = db.execute(get_user_cmd.format(user.encode('utf-8')), get_all=False)
if result:
passwd = form_data['password']
password, is_active = db.execute(get_passwd_cmd.format(user.encode('utf-8')), get_all=False)
if password == passwd and is_active == 'Y':
return True, 'ok'
elif password != passwd:
return False, 'password error'
return False, 'user not active'
return False, 'user not found'
def get_locus_result(genename, blast_results):
cds_seq_dict = blast_results.get('cds_seq', 'NA')
pro_seq_dict = blast_results.get('pro_seq', 'NA')
db = DB()
locus_result = {}
cmd = """select l.*, f.BLAST_Hit_Accession, f.Description, f.Pfam_ID,
f.Interpro_ID, f.GO_ID from locus l left join func f
on l.GENE_ID=f.GENE_ID where l.GENE_ID='{0}';
""".format(genename)
result = db.execute(cmd, get_all=False)
if result:
gene_id, chr, pos_start, pos_end = result[1:5]
blast_hit, description, pfam_id, interpro_id, go_id = result[5:]
locus_result['gene_identification'] = {
'Gene Product Name': description,
'Locus Name': genename
}
locus_result['gene_attributes'] = {
'Chromosome': chr,
"Gene Postion": '{start} - {end}'.format(start=pos_start,
end=pos_end)
}
header = [
'Accession', 'Description', 'Pfam_ID', 'Interpro_ID', 'GO_ID'
]
locus_result['gene_annotation'] = {}
locus_result['gene_annotation']['header'] = header
locus_result['gene_annotation']['body'] = [
blast_hit, description, pfam_id, interpro_id, go_id
]
locus_result['gene_cds_seq'] = cds_seq_dict
locus_result['gene_pro_seq'] = pro_seq_dict
return locus_result
def run_blast_result(genename):
# before run blast rm last search results anyway
if os.listdir(BLAST_OUT_PATH):
rm_cmd = 'rm {}'.format(os.path.join(BLAST_OUT_PATH, '*'))
subprocess.call(rm_cmd, shell=True)
# get search trans genes
db = DB()
results = db.execute(
"select GENE_TRANS from geneTrans_map where GENE='{gene}'".format(
gene=genename))
results = [result[0] for result in results]
if results:
blast_cmd = "blastdbcmd -entry {genename} -db '{db}' -line_length 100 -out {out}"
for gene_trans in results:
run_cds_cmd = blast_cmd.format(
genename=gene_trans,
db=os.path.join(BASE_DB_DIR, BLAST_CDS_DB),
out=os.path.join(BLAST_OUT_PATH, 'gene.cds.' + gene_trans))
run_protein_cmd = blast_cmd.format(
genename=gene_trans,
db=os.path.join(BASE_DB_DIR, BLAST_PROTEIN_DB),
out=os.path.join(BLAST_OUT_PATH, 'gene.protein.' + gene_trans))
subprocess.call(run_cds_cmd, shell=True)
subprocess.call(run_protein_cmd, shell=True)
blast_results = get_blast_result(results)
return blast_results
return {}
def run_snp_variations(group_info, user):
group_name = group_info.keys()
groupA = group_info[group_name[0]]
groupB = group_info[group_name[1]]
create_group_info(groupA, groupB, filename='vs'.join(group_name))
cmd = "python {script} -i {input} -o {output} -g {group} -d {depth}".format(
script=os.path.join(SNP_SCRIPT_DIR, SCRIPT_FILE),
input=INPUT_TABLE,
output=os.path.join(basedir, 'app', 'static', 'variation_results',
'vs'.join(group_name) + '_table'),
group=os.path.join(SNP_SCRIPT_DIR, 'vs'.join(group_name)),
depth='5')
subprocess.call(cmd, shell=True)
os.chdir(os.path.join(basedir, 'app', 'static', 'variation_results'))
zip_cmd = 'zip {0} {1}'.format('vs'.join(group_name) + '_table.zip',
'vs'.join(group_name) + '_table')
subprocess.call(zip_cmd, shell=True)
'''
rm_cmd = 'rm -rf {0}'.format(
'vs'.join(group_name) + '_table'
)
subprocess.call(rm_cmd, shell=True)
'''
db = DB()
results = db.execute(
"select email from user where username='******'".format(user))
if results[0][0]:
to = results[0][0]
send_mail(to,
'Snp Variation Results',
'mail/variation_results',
user=user,
filename='vs'.join(group_name) + '_table')
return 'done'
def fetch_blast_result(genename):
MAX_ROW_LEN = 125
db = DB()
command = "select GENE_ID,VAL from {table} where GENE_ID like '{gene}%'"
pep_results = db.execute(command.format(table='pep_tb', gene=wildcard_gene(genename)))
cds_results = db.execute(command.format(table='cds_tb', gene=wildcard_gene(genename)))
if len(pep_results) == 0 and len(cds_results) == 0:
return {}
pro_seq = {k:v for k,v in pep_results}
cds_seq = {k:v for k,v in cds_results}
# print it pretty
for k,v in pro_seq.items():
if len(v) > MAX_ROW_LEN:
i = 0
over_len = math.ceil(len(v) / MAX_ROW_LEN) * MAX_ROW_LEN
tmp_str = ""
while i < over_len:
tmp_str += v[i:i+MAX_ROW_LEN] + '\n'
i += MAX_ROW_LEN
pro_seq[k] = tmp_str
for k,v in cds_seq.items():
if len(v) > MAX_ROW_LEN:
i = 0
over_len = math.ceil(len(v) / MAX_ROW_LEN) * MAX_ROW_LEN
tmp_str = ""
while i < over_len:
tmp_str += v[i:i+MAX_ROW_LEN] + '\n'
i += MAX_ROW_LEN
cds_seq[k] = tmp_str
return {'pro_seq': pro_seq,
'cds_seq': cds_seq}
def batch_query_gene(genes, max_input=1000):
'''
get a gene string by search locus database
'''
if ',' in genes:
gene_list = []
genes = [each.split(',') for each in genes.split()]
for gene_part in genes:
gene_list += gene_part
else:
gene_list = genes.split()
if len(gene_list) > max_input:
return []
if len(gene_list) == 1:
_search = "('{0}')".format(gene_list[0])
else:
_search = tuple([str(each) for each in gene_list])
db = DB()
cmd = """select l.*, f.Description, f.Pfam_Description,
f.Interpro_Description, f.GO_Description from locus l left join func f
on l.GENE_ID=f.GENE_ID where l.GENE_ID in {0};
""".format(_search)
result = db.execute(cmd)
if result:
result = [(each[1],) + each[5:] for each in result]
df1 = pd.DataFrame(result)
df2 = pd.DataFrame(gene_list)
df3 = pd.merge(df2, df1, how='left').fillna("")
return [list(df3.iloc[i,:]) for i in range(len(df3))]
return []
def fetch_expression_data(gene_id, samples, table="iwgsc_refseq"):
sample_str = ','.join(samples)
cmd = "select {samples} from {table} where gene='{gene}'".format(
samples=sample_str, table=table, gene=gene_id)
db = DB()
results = db.execute(cmd)
if len(results) == 0:
return []
return list(results[0])
def get_locus_result(genename, blast_results):
cds_seq_dict = blast_results.get('cds_seq', 'NA')
pro_seq_dict = blast_results.get('pro_seq', 'NA')
db = DB()
locus_result = {}
cmd = """select l.*, f.Description, f.Pfam_Description,
f.Interpro_Description, f.GO_Description from locus l left join func f
on l.GENE_ID=f.GENE_ID where l.GENE_ID='{0}';
""".format(genename)
result = db.execute(cmd, get_all=False)
if result:
locus_result['orthologous_gene'] = {}
ortho_header = ['Arabidopsis_thaliana', 'Hordeum_vulgare', 'Oryza_sativa', 'Triticum_aestivum', 'Zea_mays']
locus_result['orthologous_gene']['header'] = ortho_header
locus_result['orthologous_gene']['body'] = []
cmd = "select l.GENE_ID, o.* from locus l left join ortho o on l.GENE_ID=o.GENE_ID where l.GENE_ID='{0}';".format(genename)
ortho_result = db.execute(cmd, get_all=False)
if ortho_result:
ortho_result_list = ortho_result[3:]
ortho_result_list = [printPretty(each) for each in ortho_result_list if each is not None]
locus_result['orthologous_gene']['body'] = ortho_result_list
gene_id, chr, pos_start, pos_end = result[1:5]
description, pfam_desc, interpro_desc, go_desc = result[5:]
locus_result['gene_identification'] = {'Gene Product Name': description,
'Locus Name': genename}
locus_result['gene_attributes'] = {'Chromosome': chr,
"Gene Postion":'{start} - {end}'.format(start=pos_start, end=pos_end)}
header = ['Description', 'Pfam_Description', 'Interpro_Description', 'GO_Description']
locus_result['gene_annotation'] = {}
locus_result['gene_annotation']['header'] = header
locus_result['gene_annotation']['body'] = [description, pfam_desc, interpro_desc, go_desc]
# match 01G and 02G TraesCS1A02G000100
#result = db.execute("select * from tissue_expression where Gene_id='{0}'".format(genename))
result = db.execute("select * from tissue_expression where Gene_id like '{0}'".format(wildcard_gene(genename)))
if result:
row = [float(each) for each in result[0][2:]]
else:
row = []
locus_result['tissue_expression'] = row
locus_result['gene_cds_seq'] = cds_seq_dict
locus_result['gene_pro_seq'] = pro_seq_dict
return locus_result
def get_expr_table(table, gene_ids, groupA, groupB, map_groupA, map_groupB):
db = DB()
select_columns = ['GENE_ID', 'CHR', 'POS_START', 'POS_END'
] + groupA + groupB
select_columns_str = ','.join(select_columns)
results = []
for gene in gene_ids:
cmd = "select {columns} from {table} where GENE_ID='{gene_id}';".format(
columns=select_columns_str, table=table, gene_id=gene)
result = db.execute(cmd, get_all=False)
if not result:
return (gene, '')
results.append(list(result))
return_select_columns = ['GENE_ID', 'CHR', 'POS_START', 'POS_END'
] + map_groupA + map_groupB
return return_select_columns, results
def clean_db_cache(table):
db = DB()
db.execute("delete from {table};".format(table=table))
def fetch_sample(table, fixed_column_num):
cmd = "select COLUMN_NAME from information_schema.COLUMNS where table_name='{table}';".format(table=table)
db = DB()
results = db.execute(cmd)
results = [each[0] for each in results]
return results[fixed_column_num:]
