def async_run_enrich(specie, genefile):
tmp_dir = randSeq(k=5)
enrich_dir = os.path.join(UPLOAD_FOLDER, 'enrich', tmp_dir)
processor.Run("mkdir {tmpDir}".format(tmpDir=enrich_dir))
if type(genefile) == type([]):
filepath = os.path.join(enrich_dir, 'tmp.gene.list')
f = open(filepath, 'w')
for each in genefile:
f.write(each + '\n')
f.close()
else:
filepath = os.path.join(enrich_dir, genefile)
processor.Run("mv {gene_file} {tmpDir}".format(gene_file=os.path.join(
UPLOAD_FOLDER, 'enrich', genefile),
tmpDir=enrich_dir))
cmd = "{pypath} {script}\
-s '{specie}' \
-g {genelist} \
-o {dir}".format(pypath=KOBAS_PATH,
script=ENRICH_SCRIPT,
specie=specie,
genelist=filepath,
dir=enrich_dir)
# print(cmd)
# print(os.environ['HOME'])
processor.shRun(cmd)
href = []
body = []
try:
f = open(os.path.join(enrich_dir, 'enrich.txt'), 'r')
head = f.readline().strip().split('\t')
head = head[:7] + head[9:]
row = f.readline()
while row:
row_list = row.strip().split('\t')
href.append(row_list[8])
body.append(row_list[:7] + row_list[9:])
row = f.readline()
# processor.Run("rm -rf {0}".format(enrich_dir))
return {
'task': 'enrich',
'result': {
'header': head,
'body': body,
'href': href
}
}
except IOError as e:
print(e)
# processor.Run("rm -rf {0}".format(enrich_dir))
return {
'task': 'enrich',
'result': {
'header': [],
'body': [],
'href': []
}
}
def run_annotation(vcf_file, annotation_database):
annotation_prefix = '-'.join([str(int(time.time())), vcf_file.split('.vcf.gz')[0]])
cmd ="{script} {vcf_file} {annotation_database} {prefix}".format(
script=vcf_ann_script,
vcf_file=os.path.join(UPLOAD_FOLDER, 'vcf_ann', vcf_file),
annotation_database=annotation_database,
prefix=os.path.join(UPLOAD_FOLDER, 'vcf_ann', annotation_prefix)
)
processor.shRun(cmd)
processor.Run("zip {zipfile} {files}".format(
zipfile=os.path.join(UPLOAD_FOLDER, 'vcf_ann', annotation_prefix + '.zip'),
files=os.path.join(UPLOAD_FOLDER, 'vcf_ann', annotation_prefix) + '.ann.vcf.*'))
return annotation_prefix + '.zip'
def fetch_vcf_samples(vcf, vcf_type="WES"):
# split vcf return each sample
fetch_sample_cmd = "sh {script} {vcf}".format(
script=extract_vcf_sample_script, vcf=vcf)
processor.shRun(fetch_sample_cmd)
f = open(os.path.join(Config.VCF_FILE_PATH, vcf + '.sample_name'), 'r')
samples = [each.strip() for each in f.readlines()]
f.close()
tc_series = tc_map(vcf, vcf_type, samples)
# split each sample
split_vcf_cmd = "sh {script} {vcf} {map_file}".format(
script=split_vcf_sample_script, vcf=vcf, map_file=vcf + '.idmap')
processor.shRun(split_vcf_cmd)
return tc_series
def async_fetch_vcf_samples(vcf, username, vcf_type="WES"):
# split vcf return each sample
fetch_sample_cmd = "sh {script} {vcf}".format(
script=extract_vcf_sample_script, vcf=vcf)
processor.shRun(fetch_sample_cmd)
f = open(os.path.join(Config.VCF_FILE_PATH, vcf + '.sample_name'), 'r')
samples = [each.strip() for each in f.readlines()]
f.close()
tc_series = tc_map(vcf, vcf_type, samples)
# creare sample in mysql
for each in tc_series:
row = Data(tc_id=each[0],
provider=username,
sample_name=each[1],
type=vcf_type)
row.save()
# split each sample
split_vcf_cmd = "sh {script} {vcf} {map_file}".format(
script=split_vcf_sample_script, vcf=vcf, map_file=vcf + '.idmap')
processor.shRun(split_vcf_cmd)
return {
'task': 'vcf_upload',
'result': '{0} upload success...'.format(vcf)
}
def snp_info(info):
tmp_param = {
'not_a_group_id': info['group'],
'group_names': ["not_a_group_id"]
}
genes = info.get('gene_list')
if genes:
gene_list = parseInput(genes)
if len(gene_list) == 0:
return {'task': 'snp_info', 'result': {'header': [], 'body': []}}
tmp_param.update({'gene_id': gene_list})
else:
tmp_param.update({
'chrom': info['chr'],
'chrom_start': info['pos_start'],
'chrom_end': info['pos_end']
})
cmd = "snpInf \
--gene_bed {gene_bed} \
--vcf_ann_file /data/wheatdb/data/vcf_private/snp.ann.table.pkl \
--vcf_dir {vcf_table} \
--outdir {outdir} \
--parameters '{param}'".format(
gene_bed=gene_bed_file,
vcf_table=Config.VCF_TABLE_PATH,
param=json.dumps(tmp_param),
outdir=ANN_PATH,
)
print(cmd)
result = processor.shRun(cmd)
if result:
head_data = result[0].split('\t')
body_data = [row.split('\t') for row in result[1:]]
return {
'task': 'snp_info',
'result': {
'header': head_data,
'body': body_data
}
}
return {'task': 'snp_info', 'result': {'header': [], 'body': []}}
def compare_info(info):
genes = info.get('gene_id')
if genes:
gene_list = parseInput(genes)
if len(gene_list) == 0:
return {
'task': 'compare_info',
'result': {
'header': [],
'body': []
}
}
info['gene_id'] = gene_list
cmd = "snpInf \
--gene_bed {gene_bed} \
--vcf_ann_file /data/wheatdb/data/vcf_private/snp.ann.table.pkl \
--vcf_dir {vcf_table} \
--outdir {outdir} \
--parameters '{param}'".format(
gene_bed=gene_bed_file,
vcf_table=Config.VCF_TABLE_PATH,
param=json.dumps(info),
outdir=ANN_PATH,
)
#print(cmd)
result = processor.shRun(cmd)
if result:
head_data = result[0].split('\t')
body_data = [row.split('\t') for row in result[1:]]
return {
'task': 'compare_info',
'result': {
'header': head_data,
'body': body_data
}
}
return {'task': 'compare_info', 'result': {'header': [], 'body': []}}
def run_bsa(info):
freq_pattern = 'snp.freq.plot.jpg'
score_pattern = 'var.score.plot.jpg'
print(info)
cmd = "snpScore -p '{info}' -d {vcf_dir} -o {out_dir} --vcf_ann_file /data/wheatdb/data/vcf_private/snp.ann.table.pkl".format(
info=info, vcf_dir=Config.VCF_TABLE_PATH, out_dir=MAPPING_PATH)
# test
print(cmd)
result = processor.shRun(cmd)
result_base = result[0]
result_path = os.path.join(result_base, 'results')
print(result_path)
processor.Run(cmd="cd {dir} && zip -r {zip_file} results".format(
zip_file=os.path.join(result_base, 'results.zip'), dir=result_base))
print("cd {dir} && zip -r {zip_file} results".format(zip_file=os.path.join(
result_base, 'results.zip'),
dir=result_base))
all_files = os.listdir(result_path)
path = result_path.split('/home/app/vcfweb/wheatdb/app')[-1]
freq_files = [
os.path.join(path, file) for file in all_files
if file[-len(freq_pattern):] == freq_pattern
]
score_files = [
os.path.join(path, file) for file in all_files
if file[-len(score_pattern):] == score_pattern
]
return {
'task': 'bsa',
'result': {
'path': os.path.join(path, '../results.zip'),
'files': freq_files + score_files
}
}