def download_species_ensembl(species, valid_species, url):
assert species in valid_species, "{0} is not in the species list".format(species)
# We want to download assembly and annotation for given species
ann_url = urljoin(url, "gtf/{0}".format(species))
cds_url = urljoin(url, "fasta/{0}/cds".format(species))
for u in (ann_url, cds_url):
valid_files = [x for x in ls_ftp(u) if x.endswith(".gz")]
for f in valid_files:
f = urljoin(u, f)
download(f)
def download_species_ensembl(species, valid_species, url):
assert species in valid_species, \
"{0} is not in the species list".format(species)
# We want to download assembly and annotation for given species
ann_url = urljoin(url, "gtf/{0}".format(species))
cds_url = urljoin(url, "fasta/{0}/cds".format(species))
for u in (ann_url, cds_url):
valid_files = [x for x in ls_ftp(u) if x.endswith(".gz")]
for f in valid_files:
f = urljoin(u, f)
download(f)
def sra(args):
"""
%prog sra term
Given an SRA run ID, fetch the corresponding .sra file
from the sra-instant FTP
"""
sra_base_url = "ftp://ftp-trace.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/"
sra_run_id_re = re.compile(r'^([DES]{1}RR)(\d{3})(\d{3,4})$')
p = OptionParser(sra.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
term, = args
m = re.search(sra_run_id_re, term)
if m is None:
logging.error("Incorrect SRA identifier format " + \
"[should be like SRR126150, SRR1001901. " + \
"len(identifier) should be between 9-10 characters]")
sys.exit()
prefix, subprefix = m.group(1), "{0}{1}".format(m.group(1), m.group(2))
download_url = urljoin(sra_base_url, prefix, subprefix, term, "{0}.sra".format(term))
logging.debug("Downloading file: {0}".format(download_url))
download(download_url)
def request(self, urn: str, params: dict):
""" Send a request to CoinMarketCap
Parameters
----------
urn : `str`
the endpoints, E.g "cryptocurrency/info"
params : `dict`
the parameters for the request
Raises
------
requests.exceptions.HTTPError
If status code is not 200
"""
url = Request("GET", urljoin(self._url, urn),
params=params).prepare().url
# NOTE: race condition, but it should be harmless
if self._session.cache.has_url(url):
response = self._request_cache(url)
else:
response = self._request_throttle(url)
res = loads(response.text)
if response.status_code == 200:
res["cached"] = response.from_cache
return res
else:
raise response.raise_for_status()
def sra(args):
"""
%prog sra term
Given an SRA run ID, fetch the corresponding .sra file
from the sra-instant FTP
"""
sra_base_url = "ftp://ftp-trace.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/"
sra_run_id_re = re.compile(r'^([DES]{1}RR)(\d{3})(\d{3,4})$')
p = OptionParser(sra.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
term, = args
m = re.search(sra_run_id_re, term)
if m is None:
logging.error("Incorrect SRA identifier format " + \
"[should be like SRR126150, SRR1001901. " + \
"len(identifier) should be between 9-10 characters]")
sys.exit()
prefix, subprefix = m.group(1), "{0}{1}".format(m.group(1), m.group(2))
download_url = urljoin(sra_base_url, prefix, subprefix, term,
"{0}.sra".format(term))
logging.debug("Downloading file: {0}".format(download_url))
download(download_url)
def get_features(refseq, start, end, strand, featuretype, completely_within=True, level=1):
"""
Return the features within a given set of chromosome coordinates
"""
url = urljoin(THALEMINE_BASE_URL, "jbrowse", TAXID, "features", refseq)
data = tools.do_request(url, None, start=start, end=end, type=featuretype)
elems_to_delete = []
for x, elem0 in enumerate(data['features']):
# remove feature if not completely_within specified chromosome coordinates
if completely_within and (elem0['start'] < start or elem0['end'] > end):
elems_to_delete.append(x)
continue
# remove all subfeatures below 0th-level object
if level == 0:
data['features'][x]['subfeatures'] = []
# remove all subfeatures below 1st-level object
elif level == 1:
for y, elem1 in enumerate(elem0['subfeatures']):
data['features'][x]['subfeatures'][y]['subfeatures'] = []
for i in sorted(elems_to_delete, reverse=True):
del data['features'][i]
return data
def get_global_stats(featuretype):
"""
Return global stats for features of specific type
"""
url = urljoin(THALEMINE_BASE_URL, "jbrowse", TAXID, "stats", "global")
global_stats = tools.do_request(url, None, type=featuretype)
return global_stats
def get_region_feature_densities(refseq, start, end, featuretype):
"""
Return binned density stats for features within a given set of chromosome coordinates
"""
url = urljoin(THALEMINE_BASE_URL, "jbrowse", TAXID, "stats", "regionFeatureDensities", refseq)
region_feature_densities = tools.do_request(url, None, start=start, end=end, type=featuretype)
return region_feature_densities
def download_species_phytozome9(species, valid_species, base_url, assembly=False):
assert species in valid_species, "{} is not in the species list".format(species)
# We want to download assembly and annotation for given species
surl = urljoin(base_url, species)
contents = [x for x in ls_ftp(surl) if x.endswith("_readme.txt")]
magic = contents[0].split("_")[1] # Get the magic number
logging.debug("Found magic number for {0}: {1}".format(species, magic))
pf = "{0}_{1}".format(species, magic)
asm_url = urljoin(surl, "assembly/{0}.fa.gz".format(pf))
ann_url = urljoin(surl, "annotation/{0}_gene.gff3.gz".format(pf))
cds_url = urljoin(surl, "annotation/{0}_cds.fa.gz".format(pf))
res = {}
if assembly:
res["asm"] = download(asm_url)
res["gff"] = download(ann_url)
res["cds"] = download(cds_url)
return res
def do(self):
"run it, get a new url"
scheme, netloc, path, params, query, fragment = Split(self.url).do()
if isinstance(self.query, dict):
query = query + "&" + urllib.urlencode(self.query) if query else urllib.urlencode(self.query)
path = urljoin(path, self.path).replace('\\', '/') if self.path else path
return Splice(scheme=scheme, netloc=netloc, path=path, params=params, query=query, fragment=fragment).geturl
def download_species_phytozome(species, valid_species, url, assembly=False):
from os.path import join as urljoin
assert species in valid_species, \
"{0} is not in the species list".format(species)
# We want to download assembly and annotation for given species
surl = urljoin(url, species)
contents = [x for x in ls_ftp(surl) if x.endswith("_readme.txt")]
magic = contents[0].split("_")[1] # Get the magic number
logging.debug("Found magic number for {0}: {1}".format(species, magic))
pf = "{0}_{1}".format(species, magic)
asm_url = urljoin(surl, "assembly/{0}.fa.gz".format(pf))
ann_url = urljoin(surl, "annotation/{0}_gene.gff3.gz".format(pf))
cds_url = urljoin(surl, "annotation/{0}_cds.fa.gz".format(pf))
if assembly:
download(asm_url)
for u in (ann_url, cds_url):
download(u)
def download_species_phytozome(species, valid_species, url, assembly=False):
from os.path import join as urljoin
assert species in valid_species, \
"{0} is not in the species list".format(species)
# We want to download assembly and annotation for given species
surl = urljoin(url, species)
contents = [x for x in ls_ftp(surl) if x.endswith("_readme.txt")]
magic = contents[0].split("_")[1] # Get the magic number
logging.debug("Found magic number for {0}: {1}".format(species, magic))
pf = "{0}_{1}".format(species, magic)
asm_url = urljoin(surl, "assembly/{0}.fa.gz".format(pf))
ann_url = urljoin(surl, "annotation/{0}_gene.gff3.gz".format(pf))
cds_url = urljoin(surl, "annotation/{0}_cds.fa.gz".format(pf))
if assembly:
download(asm_url)
for u in (ann_url, cds_url):
download(u)
def download_srr_term(term):
sra_base_url = "ftp://ftp-trace.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/"
sra_run_id_re = re.compile(r'^([DES]{1}RR)(\d{3})(\d{3,4})$')
m = re.search(sra_run_id_re, term)
if m is None:
logging.error("Incorrect SRA identifier format " + \
"[should be like SRR126150, SRR1001901. " + \
"len(identifier) should be between 9-10 characters]")
sys.exit()
prefix, subprefix = m.group(1), "{0}{1}".format(m.group(1), m.group(2))
download_url = urljoin(sra_base_url, prefix, subprefix, term, "{0}.sra".format(term))
logging.debug("Downloading file: {0}".format(download_url))
return download(download_url)
def download_srr_term(term):
sra_base_url = "ftp://ftp-trace.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/"
sra_run_id_re = re.compile(r'^([DES]{1}RR)(\d{3})(\d{3,4})$')
m = re.search(sra_run_id_re, term)
if m is None:
logging.error("Incorrect SRA identifier format " + \
"[should be like SRR126150, SRR1001901. " + \
"len(identifier) should be between 9-10 characters]")
sys.exit()
prefix, subprefix = m.group(1), "{0}{1}".format(m.group(1), m.group(2))
download_url = urljoin(sra_base_url, prefix, subprefix, term, "{0}.sra".format(term))
logging.debug("Downloading file: {0}".format(download_url))
return download(download_url)
def rule(services, settings):
if not settings.ONE_DOMAIN_MODE:
return
API_URL = urljoin(settings.API_URL, "api")
STORE_AVAILABLE = services.get("store")
ADMIN_AVAILABLE = services.get("admin")
BACKEND_AVAILABLE = services.get("backend")
# replace defaults
if STORE_AVAILABLE:
with modify_key(services, "store", "environment") as environment:
environment["BITCART_STORE_API_URL"] = API_URL
if ADMIN_AVAILABLE:
with modify_key(services, "admin", "environment") as environment:
environment["BITCART_ADMIN_ROOTPATH"] = environment[
"BITCART_ADMIN_ROOTPATH"].replace("/", "/admin")
environment["BITCART_ADMIN_API_URL"] = API_URL
elif ADMIN_AVAILABLE:
with modify_key(services, "admin", "environment") as environment:
environment["BITCART_ADMIN_API_URL"] = API_URL
if BACKEND_AVAILABLE and (ADMIN_AVAILABLE or STORE_AVAILABLE):
with modify_key(services, "backend", "environment") as environment:
environment["BITCART_BACKEND_ROOTPATH"] = environment[
"BITCART_BACKEND_ROOTPATH"].replace("-}", "-/api}")
def entrez(args):
"""
%prog entrez <filename|term>
`filename` contains a list of terms to search. Or just one term. If the
results are small in size, e.g. "--format=acc", use "--batchsize=100" to speed
the download.
"""
p = OptionParser(entrez.__doc__)
allowed_databases = {
"fasta": ["genome", "nuccore", "nucgss", "protein", "nucest"],
"asn.1": ["genome", "nuccore", "nucgss", "protein", "gene"],
"xml": ["genome", "nuccore", "nucgss", "nucest", "gene"],
"gb": ["genome", "nuccore", "nucgss"],
"est": ["nucest"],
"gss": ["nucgss"],
"acc": ["nuccore"],
}
valid_formats = tuple(allowed_databases.keys())
valid_databases = ("genome", "nuccore", "nucest", "nucgss", "protein", "gene")
p.add_option(
"--noversion",
dest="noversion",
default=False,
action="store_true",
help="Remove trailing accession versions",
)
p.add_option(
"--format",
default="fasta",
choices=valid_formats,
help="download format",
)
p.add_option(
"--database",
default="nuccore",
choices=valid_databases,
help="search database",
)
p.add_option(
"--retmax",
default=1000000,
type="int",
help="how many results to return",
)
p.add_option(
"--skipcheck",
default=False,
action="store_true",
help="turn off prompt to check file existence",
)
p.add_option(
"--batchsize",
default=500,
type="int",
help="download the results in batch for speed-up",
)
p.set_outdir(outdir=None)
p.add_option("--outprefix", default="out", help="output file name prefix")
p.set_email()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(p.print_help())
(filename,) = args
if op.exists(filename):
pf = filename.rsplit(".", 1)[0]
list_of_terms = [row.strip() for row in open(filename)]
if opts.noversion:
list_of_terms = [x.rsplit(".", 1)[0] for x in list_of_terms]
else:
pf = filename
# the filename is the search term
list_of_terms = [filename.strip()]
fmt = opts.format
database = opts.database
batchsize = opts.batchsize
assert (
database in allowed_databases[fmt]
), "For output format '{0}', allowed databases are: {1}".format(
fmt, allowed_databases[fmt]
)
assert batchsize >= 1, "batchsize must >= 1"
if " " in pf:
pf = opts.outprefix
outfile = "{0}.{1}".format(pf, fmt)
outdir = opts.outdir
if outdir:
mkdir(outdir)
# If noprompt, will not check file existence
if not outdir:
fw = must_open(outfile, "w", checkexists=True, skipcheck=opts.skipcheck)
if fw is None:
return
seen = set()
totalsize = 0
for id, size, term, handle in batch_entrez(
list_of_terms,
retmax=opts.retmax,
rettype=fmt,
db=database,
batchsize=batchsize,
email=opts.email,
):
if outdir:
outfile = urljoin(outdir, "{0}.{1}".format(term, fmt))
fw = must_open(outfile, "w", checkexists=True, skipcheck=opts.skipcheck)
if fw is None:
continue
rec = handle.read()
if id in seen:
logging.error("Duplicate key ({0}) found".format(rec))
continue
totalsize += size
print(rec, file=fw)
print(file=fw)
seen.add(id)
if seen:
printf(
"A total of {0} {1} records downloaded.".format(totalsize, fmt.upper()),
)
return outfile
def search(args):
"""
args contains a dict with one or key:values
transcript is AGI identifier and is mandatory
material is the tissue or treatment and is restricted below to a limited list
"""
"""
In the future, ADAMA will check a query, map_*, or generic request against a list of mandatory
parameters specified for each service. For now, if we want to enforce that behavior we need to
implement it ourselves.
ADAMA will have a graceful, cross-language exception handling scheme in a future release
At present, we are hand-coding a return
"""
if not args.viewkeys() & {'material1', 'material2', 'foldchange'}:
return
"""
Check that foldchange is a valid number
"""
foldchange = args['foldchange']
try:
n = float(foldchange)
except (ValueError, TypeError):
return
"""
Check materials to make sure they're in the (hard-coded) approved list
"""
valid_materials = { 'flower': 'Flo', 'iaa': 'IAA', 'leaf': 'Lea', \
'root': 'Roo', 'salicylic': 'Sal', 'nacl': 'NaC', \
'young': 'You', 't87': 'T87'}
material1 = args['material1'].lower()
if material1 not in valid_materials.keys():
return
tissue1 = valid_materials[material1]
material2 = args['material2'].lower()
if material2 not in valid_materials.keys():
return
tissue2 = valid_materials[material2]
"""
Build the url from the base + the intended endpoint action
Also encode the params (payload) into a dict
"""
url = urljoin(jcvi_common.base_url(), 'ExpressionConditionComparison')
payload = { 'tissue1': tissue1, 'tissue2': tissue2, 'change': foldchange }
"""
Make the request to the remote service
"""
r = requests.get(url, params=payload)
"""
Iterate through the results
Foreach record from the remote service, build the response json
Print this json to stdout followed by a record separator "---"
ADAMA takes care of serializing these results
"""
p = re.compile('AT[1-5MC]G[0-9]{5,5}\.[0-9]+', re.IGNORECASE)
for result in r.json()['compare_table']:
# check that transcript uses a valid transcript identifier
transcript = result['elem_target_id']
if not p.search(transcript):
continue
record = {
'transcript': transcript,
'class': 'transcript_property',
'source_text_description': 'RT-PCR',
'expression_comparison_record': {
'material1_text_description': result['elem_tissue1'],
'expression_value_material1': result['elem_tissue1_value'],
'expression_value_material1_stdev': result['elem_tissue1_value2'],
'material2_text_description': result['elem_tissue2'],
'expression_value_material2': result['elem_tissue2_value'],
'expression_value_material2_stdev': result['elem_tissue2_value2']
}
}
print json.dumps(record, indent=2)
print '---'
def __init__(self, request, endpoint):
self.request = lambda x: request(urljoin(endpoint, "price-conversion"),
args(**x))
def __init__(self, request, endpoint):
self.request = lambda x: request(urljoin(endpoint, "map"), x)
def __init__(self, request, endpoint):
self.request = lambda x, y: request(urljoin(endpoint, "quotes", x),
args(**y))
def __init__(self, request, endpoint):
self.request = lambda x: request(
urljoin(endpoint, "market-pairs/latest"), args(**x))
def __init__(self, request, endpoint):
self.request = lambda x: request(urljoin(endpoint, "info"), args(**x))
def entrez(args):
"""
%prog entrez <filename|term>
`filename` contains a list of terms to search. Or just one term. If the
results are small in size, e.g. "--format=acc", use "--batchsize=100" to speed
the download.
"""
p = OptionParser(entrez.__doc__)
allowed_databases = {"fasta": ["genome", "nuccore", "nucgss", "protein", "nucest"],
"asn.1": ["genome", "nuccore", "nucgss", "protein"],
"gb" : ["genome", "nuccore", "nucgss"],
"est" : ["nucest"],
"gss" : ["nucgss"],
"acc" : ["nuccore"],
}
valid_formats = tuple(allowed_databases.keys())
valid_databases = ("genome", "nuccore", "nucest", "nucgss", "protein")
p.add_option("--noversion", dest="noversion",
default=False, action="store_true",
help="Remove trailing accession versions")
p.add_option("--format", default="fasta", choices=valid_formats,
help="download format [default: %default]")
p.add_option("--database", default="nuccore", choices=valid_databases,
help="search database [default: %default]")
p.add_option("--retmax", default=1000000, type="int",
help="how many results to return [default: %default]")
p.add_option("--skipcheck", default=False, action="store_true",
help="turn off prompt to check file existence [default: %default]")
p.add_option("--batchsize", default=500, type="int",
help="download the results in batch for speed-up [default: %default]")
p.add_option("--outdir", default=None,
help="output directory, with accession number as filename")
p.add_option("--outprefix", default="out",
help="output file name prefix [default: %default]")
p.set_email()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(p.print_help())
filename, = args
if op.exists(filename):
pf = filename.rsplit(".", 1)[0]
list_of_terms = [row.strip() for row in open(filename)]
if opts.noversion:
list_of_terms = [x.rsplit(".", 1)[0] for x in list_of_terms]
else:
pf = filename
# the filename is the search term
list_of_terms = [filename.strip()]
fmt = opts.format
database = opts.database
batchsize = opts.batchsize
assert database in allowed_databases[fmt], \
"For output format '{0}', allowed databases are: {1}".\
format(fmt, allowed_databases[fmt])
assert batchsize >= 1, "batchsize must >= 1"
if " " in pf:
pf = opts.outprefix
outfile = "{0}.{1}".format(pf, fmt)
outdir = opts.outdir
if outdir:
mkdir(outdir)
# If noprompt, will not check file existence
if not outdir:
fw = must_open(outfile, "w", checkexists=True, \
skipcheck=opts.skipcheck)
if fw is None:
return
seen = set()
totalsize = 0
for id, size, term, handle in batch_entrez(list_of_terms, retmax=opts.retmax, \
rettype=fmt, db=database, batchsize=batchsize, \
email=opts.email):
if outdir:
outfile = urljoin(outdir, "{0}.{1}".format(term, fmt))
fw = must_open(outfile, "w", checkexists=True, \
skipcheck=opts.skipcheck)
if fw is None:
continue
rec = handle.read()
if id in seen:
logging.error("Duplicate key ({0}) found".format(rec))
continue
totalsize += size
print >> fw, rec
print >> fw
seen.add(id)
if seen:
print >> sys.stderr, "A total of {0} {1} records downloaded.".\
format(totalsize, fmt.upper())
return outfile
def urljoin(url, path):
return path.urljoin(url, path)
