def gene_network2ws(meth, feature_set_id=None, out_id=None):
""" Query all available network data in KBase central store.
:param feature_set_id: FeatureSet workspace object id
:type feature_set_id: kbtypes.KBaseSearch.FeatureSet
:param out_id: Output Networks object identifier
:type out_id: kbtypes.KBaseNetworks.Network
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
meth.advance("Retrieve genes from workspace")
ws = Workspace2(token=meth.token, wsid=meth.workspace_id)
fs = ws.get(feature_set_id)
qid2cds = ids2cds(fs['elements'].keys())
gl_str = ",".join(list(set(qid2cds.values())));
meth.advance("Running GeneList to Networks")
argsx = {"ws_id" : meth.workspace_id, "inobj_id" : gl_str, "outobj_id": out_id}
try:
gl_oid = gc.genelist_to_networks(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
meth.advance("Returning object")
return _workspace_output(out_id)
def gwas_run_gwas2(meth, genotype=None, kinship_matrix=None, traits=None, out=None):
"""Computes association between each SNP and a trait of interest that has been scored
across a large number of individuals. This method takes Filtered SNP object,
kinship matrix, trait object as input and computes association.
:param genotype: Population variation object
:type genotype: kbtypes.KBaseGwasData.GwasPopulationVariation
:param kinship_matrix: Kinship matrix object id
:type kinship_matrix: kbtypes.KBaseGwasData.GwasPopulationKinship
:param traits: Trait object id
:type traits: kbtypes.KBaseGwasData.GwasPopulationTrait
:param out: Output
:type out: kbtypes.KBaseGwasData.GwasTopVariations
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
argsx = {"ws_id" : meth.workspace_id, "variation_id" : genotype, "trait_id" : traits, "kinship_id": kinship_matrix, "out_id" : out, "comment" : "comment"}
meth.advance("submit job to run GWAS analysis")
try:
jid = gc.run_gwas(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
if not jid:
raise GWASException(2, "submit job failed, no job id")
AweJob.URL = URLS.awe
AweJob(meth, started="GWAS analysis using emma", running="GWAS analysis using emma").run(jid[0])
return _workspace_output(out)
def gwas_run_kinship(meth, filtered_variation=None, out=None, comment=None):
"""Computes the n by n kinship matrix for a set of n related subjects.
The kinship matrix defines pairwise genetic relatedness among individuals and
is estimated by using all genotyped markers. This requires the filtered SNPs as input.
:param filtered_variation: Population variation, filtered
:type filtered_variation: kbtypes.KBaseGwasData.GwasPopulationVariation
:param out: Computed Kinship matrix
:type out: kbtypes.KBaseGwasData.GwasPopulationKinship
:param comment: Comment
:type comment: kbtypes.Unicode
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
argsx = {"ws_id" : meth.workspace_id, "inobj_id" : filtered_variation, "outobj_id" : out, "comment" : "comment"}
meth.advance("submit job to select_random_snps")
try:
jid = gc.calculate_kinship_matrix(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
if not jid:
raise GWASException(2, "submit job failed, no job id")
AweJob.URL = URLS.awe
AweJob(meth, started="Calculate Kinship matrix", running="Kinship matrix").run(jid[0])
return _workspace_output(out)
def maf(meth, maf=0.05, variation=None, out=None, comment=None):
"""Perform filtering on Minor allele frequency (MAF).
Minor allele frequency (MAF) refers to the frequency at which the least common
<a href="http://en.wikipedia.org/wiki/Allele">allele</a> occurs in a given population.
:param maf: Minor allele frequency
:type maf: kbtypes.Numeric
:param variation: Population variation object
:type variation: kbtypes.KBaseGwasData.GwasPopulationVariation
:param out: Population variation, filtered
:type out: kbtypes.KBaseGwasData.GwasPopulationVariation
:param comment: Comment
:type comment: kbtypes.Unicode
:return: Workspace ID of filtered data
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
argsx = {"ws_id" : meth.workspace_id, "inobj_id" : variation, "outobj_id" : out, "minor_allele_frequency" : maf, "comment" : "comment"}
meth.advance("submit job to filter VCF")
try:
jid = gc.prepare_variation(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
if not jid:
raise GWASException(2, "submit job failed, no job id")
AweJob.URL = URLS.awe
AweJob(meth, started="run VCF", running="VCF").run(jid[0])
return _workspace_output(out)
def gwas_variation_to_genes(meth, workspaceID=None, gwasObjectID=None, num2snps=None, pmin=None, distance=None, gl_out=None, fs_out=None):
"""This method takes the top SNPs obtained after GWAS analysis as input
(TopVariations) object, -log (pvalue) cutoff and a distance parameter as input.
For each significant SNP that passes the p-value cutoff, genes are searched in the
window specified by the distance parameter.
:param workspaceID: Workspace (use current if empty)
:type workspaceID: kbtypes.Unicode
:param gwasObjectID: GWAS analysis MLM result object
:type gwasObjectID: kbtypes.KBaseGwasData.GwasTopVariations
:param num2snps: Number to snps
:type num2snps: kbtypes.Numeric
:default num2snps: 100
:param pmin: Minimum pvalue (-log10)
:type pmin: kbtypes.Numeric
:default pmin: 4
:param distance: Distance in bp around SNP to look for genes
:type distance: kbtypes.Numeric
:default distance: 10000
:param gl_out: Output GwasGeneLint workspace object name
:type gl_out: kbtypes.KBaseGwasData.GwasGeneList
:param fs_out: Output FeatureSet workspace object name
:type fs_out: kbtypes.KBaseSearch.FeatureSet
:return: Workspace objectID of gwas results
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
if not workspaceID:
workspaceID = meth.workspace_id
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
meth.advance("Running Variations to Genes")
argsx = {"ws_id" : meth.workspace_id, "variation_id" : gwasObjectID, "out_id": gl_out, "num2snps" : num2snps, "pmin": pmin, "distance" : distance, "comment" : "comment"}
try:
gl_oid = gc.variations_to_genes(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
ws = Workspace2(token=meth.token, wsid=meth.workspace_id)
raw_data = ws.get(gl_out)
gl = [ gr[2] for gr in raw_data['genes']]
fs = genelist2fs(gl)
ws.save_objects({'workspace' : meth.workspace_id, 'objects' :[{'type' : 'KBaseSearch.FeatureSet', 'data' : fs, 'name' : fs_out, 'meta' : {'original' : gl_out}}]})
meth.advance("Returning object")
return json.dumps({'values': [
["Workspace GwasGeneList object", gl_out],
["Workspace FeatureSet object", fs_out]
]})
def gwas_run_gwas2(meth,
genotype=None,
kinship_matrix=None,
traits=None,
out=None):
"""Computes association between each SNP and a trait of interest that has been scored
across a large number of individuals. This method takes Filtered SNP object,
kinship matrix, trait object as input and computes association.
:param genotype: Population variation object
:type genotype: kbtypes.KBaseGwasData.GwasPopulationVariation
:param kinship_matrix: Kinship matrix object id
:type kinship_matrix: kbtypes.KBaseGwasData.GwasPopulationKinship
:param traits: Trait object id
:type traits: kbtypes.KBaseGwasData.GwasPopulationTrait
:param out: Output
:type out: kbtypes.KBaseGwasData.GwasTopVariations
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
argsx = {
"ws_id": meth.workspace_id,
"variation_id": genotype,
"trait_id": traits,
"kinship_id": kinship_matrix,
"out_id": out,
"comment": "comment"
}
meth.advance("submit job to run GWAS analysis")
try:
jid = gc.run_gwas(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
if not jid:
raise GWASException(2, "submit job failed, no job id")
AweJob.URL = URLS.awe
AweJob(meth,
started="GWAS analysis using emma",
running="GWAS analysis using emma").run(jid[0])
return _workspace_output(out)
def gene_network2ws(meth, obj_id=None, out_id=None):
"""This method displays a gene list
along with functional annotation in a table.
:param obj_id: Gene List workspace object identifier.
:type obj_id: kbtypes.KBaseGwasData.GwasGeneList
:param out_id: Output Networks object identifier
:type out_id: kbtypes.KBaseNetworks.Network
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
# :param workspace_id: Workspace name (if empty, defaults to current workspace)
# :type workspace_id: kbtypes.Unicode
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
meth.advance("Retrieve genes from workspace")
# if not workspace_id:
# meth.debug("Workspace ID is empty, setting to current ({})".format(meth.workspace_id))
# workspace_id = meth.workspace_id
ws = Workspace2(token=meth.token, wsid=meth.workspace_id)
raw_data = ws.get(obj_id)
gl = [gr[2] for gr in raw_data['genes']]
gl_str = ",".join(gl)
meth.advance("Running GeneList to Networks")
argsx = {
"ws_id": meth.workspace_id,
"inobj_id": gl_str,
"outobj_id": out_id
}
try:
gl_oid = gc.genelist_to_networks(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
#if not gl_oid: # it may return empty string based on current script
# raise GWASException(2, "submit job failed, no job id")
meth.advance("Returning object")
return _workspace_output(out_id)
def gene_network2ws(meth, obj_id=None, out_id=None):
"""This method displays a gene list
along with functional annotation in a table.
:param obj_id: Gene List workspace object identifier.
:type obj_id: kbtypes.KBaseGwasData.GwasGeneList
:param out_id: Output Networks object identifier
:type out_id: kbtypes.KBaseNetworks.Network
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
# :param workspace_id: Workspace name (if empty, defaults to current workspace)
# :type workspace_id: kbtypes.Unicode
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
meth.advance("Retrieve genes from workspace")
# if not workspace_id:
# meth.debug("Workspace ID is empty, setting to current ({})".format(meth.workspace_id))
# workspace_id = meth.workspace_id
ws = Workspace2(token=meth.token, wsid=meth.workspace_id)
raw_data = ws.get(obj_id)
gl = [ gr[2] for gr in raw_data['genes']]
gl_str = ",".join(gl);
meth.advance("Running GeneList to Networks")
argsx = {"ws_id" : meth.workspace_id, "inobj_id" : gl_str, "outobj_id": out_id}
try:
gl_oid = gc.genelist_to_networks(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
#if not gl_oid: # it may return empty string based on current script
# raise GWASException(2, "submit job failed, no job id")
meth.advance("Returning object")
return _workspace_output(out_id)
def maf(meth, maf=0.05, variation=None, out=None, comment=None):
"""Perform filtering on Minor allele frequency (MAF).
Minor allele frequency (MAF) refers to the frequency at which the least common
<a href="http://en.wikipedia.org/wiki/Allele">allele</a> occurs in a given population.
:param maf: Minor allele frequency
:type maf: kbtypes.Numeric
:param variation: Population variation object
:type variation: kbtypes.KBaseGwasData.GwasPopulationVariation
:param out: Population variation, filtered
:type out: kbtypes.KBaseGwasData.GwasPopulationVariation
:param comment: Comment
:type comment: kbtypes.Unicode
:return: Workspace ID of filtered data
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
argsx = {
"ws_id": meth.workspace_id,
"inobj_id": variation,
"outobj_id": out,
"minor_allele_frequency": maf,
"comment": "comment"
}
meth.advance("submit job to filter VCF")
try:
jid = gc.prepare_variation(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
if not jid:
raise GWASException(2, "submit job failed, no job id")
AweJob.URL = URLS.awe
AweJob(meth, started="run VCF", running="VCF").run(jid[0])
return _workspace_output(out)
def gwas_run_kinship(meth, filtered_variation=None, out=None, comment=None):
"""Computes the n by n kinship matrix for a set of n related subjects.
The kinship matrix defines pairwise genetic relatedness among individuals and
is estimated by using all genotyped markers. This requires the filtered SNPs as input.
:param filtered_variation: Population variation, filtered
:type filtered_variation: kbtypes.KBaseGwasData.GwasPopulationVariation
:param out: Computed Kinship matrix
:type out: kbtypes.KBaseGwasData.GwasPopulationKinship
:param comment: Comment
:type comment: kbtypes.Unicode
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
argsx = {
"ws_id": meth.workspace_id,
"inobj_id": filtered_variation,
"outobj_id": out,
"comment": "comment"
}
meth.advance("submit job to select_random_snps")
try:
jid = gc.calculate_kinship_matrix(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
if not jid:
raise GWASException(2, "submit job failed, no job id")
AweJob.URL = URLS.awe
AweJob(meth, started="Calculate Kinship matrix",
running="Kinship matrix").run(jid[0])
return _workspace_output(out)
def gene_network2ws(meth, feature_set_id=None, out_id=None):
""" Query all available network data in KBase central store.
:param feature_set_id: FeatureSet workspace object id
:type feature_set_id: kbtypes.KBaseSearch.FeatureSet
:param out_id: Output Networks object identifier
:type out_id: kbtypes.KBaseNetworks.Network
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
meth.advance("Retrieve genes from workspace")
ws = Workspace2(token=meth.token, wsid=meth.workspace_id)
fs = ws.get(feature_set_id)
qid2cds = ids2cds(fs['elements'].keys())
gl_str = ",".join(list(set(qid2cds.values())))
meth.advance("Running GeneList to Networks")
argsx = {
"ws_id": meth.workspace_id,
"inobj_id": gl_str,
"outobj_id": out_id
}
try:
gl_oid = gc.genelist_to_networks(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
meth.advance("Returning object")
return _workspace_output(out_id)
def go_enrch_net(meth, net_obj_id=None, p_value=0.05, ec=None, domain=None):
"""Identify Gene Ontology terms enriched in individual network clusters
:param net_obj_id: Cluster object id
:type net_obj_id: kbtypes.KBaseNetworks.Network
:param p_value: p-value cutoff
:type p_value: kbtypes.Unicode
:param ec: Evidence code list (comma separated, IEA, ...)
:type ec:kbtypes.Unicode
:param domain: Domain list (comma separated, biological_process, ...)
:type domain: kbtypes.Unicode
:return: Workspace id
:rtype: kbtypes.Unicode
:output_widget: GeneTableWidget
"""
# :default p_value: 0.05
meth.stages = 3
meth.advance("Prepare enrichment test")
gc = GWAS(URLS.gwas, token=meth.token)
ec = ec.replace(" ", "")
domain = domain.replace(" ", "")
ec_list = [i for i in ec.split(',')]
domain_list = [i for i in domain.split(',')]
wsd = Workspace2(token=meth.token, wsid=meth.workspace_id)
oc = Ontology(url=URLS.ontology)
net_object = wsd.get_objects([{
'workspace': meth.workspace_id,
'name': net_obj_id
}])
nc = Node(net_object[0]['data']['nodes'], net_object[0]['data']['edges'])
gids = [
i for i in sorted(nc.ugids.keys()) if 'CDS' in i or 'locus' in i or (
not 'clst' in i and not i.startswith('cluster'))
]
gids2cds = ids2cds(gids)
meth.advance("Run enrichment test for each clusters")
rows = []
for hr_nd in net_object[0]['data']['nodes']:
gid = hr_nd['entity_id']
if not (gid.startswith('cluster.') or 'clst' in gid): continue
glist = nc.get_gene_list(gid)
# now everything is in CDS id
#llist = []
#for i in glist:
# if i in mrnas: i = mrnas[i]
# if i in locus: i = locus[i]
# if 'locus' in i: llist.append(i)
#llist = [ locus[mrnas[i]] ]; # it will ignore orignal locus ids (TODO: keep locus)
cds_gl = [gids2cds[i] for i in glist]
enr_list = oc.get_go_enrichment(cds_gl, domain_list, ec_list,
'hypergeometric', 'GO')
enr_list = sorted(enr_list, key=itemgetter('pvalue'), reverse=False)
go_enr_smry = ""
go_enr_anns = ["", "", ""]
for i in range(len(enr_list)):
goen = enr_list[i]
if goen['pvalue'] > float(p_value): continue
hr_nd['user_annotations']['gce.' + goen['goID'] +
".desc"] = goen['goDesc'][0]
hr_nd['user_annotations']['gce.' + goen['goID'] +
".domain"] = goen['goDesc'][1]
hr_nd['user_annotations']['gce.' + goen['goID'] +
".p_value"] = ` goen['pvalue'] `
if i < 3:
go_enr_smry += goen['goID'] + "(" + "{:6.4f}".format(
goen['pvalue']) + ")" + goen['goDesc'][0] + "\n"
go_enr_anns[i] = goen['goID'] + "(" + "{:6.4f}".format(
goen['pvalue']) + ")" + goen['goDesc'][0]
hr_nd['user_annotations']['go_enrichnment_annotation'] = go_enr_smry
rows.append(
[gid,
len(glist), go_enr_anns[0], go_enr_anns[1], go_enr_anns[2]])
wsd.save_objects({
'workspace':
meth.workspace_id,
'objects': [{
'type': 'KBaseNetworks.Network',
'data': net_object[0]['data'],
'name': net_obj_id + ".cenr",
'meta': {
'orginal': net_obj_id
}
}]
})
rows = sorted(rows, key=lambda x: x[1], reverse=True)
#meth.debug("rows: {}".format(rows))
header = [
"Cluster ID", "# of Genes", "Annotation1", "Annotation2", "Annotation3"
]
data = {'table': [header] + rows}
return json.dumps(data)
def gwas_variation_to_genes(meth,
workspaceID=None,
gwasObjectID=None,
num2snps=None,
pmin=None,
distance=None,
gl_out=None,
fs_out=None):
"""This method takes the top SNPs obtained after GWAS analysis as input
(TopVariations) object, -log (pvalue) cutoff and a distance parameter as input.
For each significant SNP that passes the p-value cutoff, genes are searched in the
window specified by the distance parameter.
:param workspaceID: Workspace (use current if empty)
:type workspaceID: kbtypes.Unicode
:param gwasObjectID: GWAS analysis MLM result object
:type gwasObjectID: kbtypes.KBaseGwasData.GwasTopVariations
:param num2snps: Number to snps
:type num2snps: kbtypes.Numeric
:default num2snps: 100
:param pmin: Minimum pvalue (-log10)
:type pmin: kbtypes.Numeric
:default pmin: 4
:param distance: Distance in bp around SNP to look for genes
:type distance: kbtypes.Numeric
:default distance: 10000
:param gl_out: Output GwasGeneLint workspace object name
:type gl_out: kbtypes.KBaseGwasData.GwasGeneList
:param fs_out: Output FeatureSet workspace object name
:type fs_out: kbtypes.KBaseSearch.FeatureSet
:return: Workspace objectID of gwas results
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 3
if not workspaceID:
workspaceID = meth.workspace_id
meth.advance("init GWAS service")
gc = GWAS(URLS.gwas, token=meth.token)
meth.advance("Running Variations to Genes")
argsx = {
"ws_id": meth.workspace_id,
"variation_id": gwasObjectID,
"out_id": gl_out,
"num2snps": num2snps,
"pmin": pmin,
"distance": distance,
"comment": "comment"
}
try:
gl_oid = gc.variations_to_genes(argsx)
except Exception as err:
raise GWASException("submit job failed: {}".format(err))
ws = Workspace2(token=meth.token, wsid=meth.workspace_id)
raw_data = ws.get(gl_out)
gl = [gr[2] for gr in raw_data['genes']]
fs = genelist2fs(gl)
ws.save_objects({
'workspace':
meth.workspace_id,
'objects': [{
'type': 'KBaseSearch.FeatureSet',
'data': fs,
'name': fs_out,
'meta': {
'original': gl_out
}
}]
})
meth.advance("Returning object")
return json.dumps({
'values': [["Workspace GwasGeneList object", gl_out],
["Workspace FeatureSet object", fs_out]]
})
