def main():
args = parser.parse_args()
beta = 0.5
#get mutational data
mvals = getMutationalData()
##load up interactome
gfile = args.graph
##TODO: replace this with Docker image call
g = pickle.load(open(gfile, 'rb'))
key = 'mpnstPDXmuts'
this_hyp = hyphalNetwork(mvals, g)
this_hyp._to_file(key + '_hypha.pkl')
##read from file
###this is all we need to do in a single eval, then we can do tests later
this_hyp.node_stats().to_csv(key + '_nodelist.csv')
for_e = hyEnrich.go_enrich_forests(this_hyp)
this_hyp.assign_enrichment(for_e, type='forest')
for_e.to_csv(key + 'enrichedForestGoTerms.csv')
com_e = hyEnrich.go_enrich_communities(this_hyp)
this_hyp.assign_enrichment(com_e, type='community')
this_hyp._to_file(key + '_hypha.pkl')
com_e.to_csv(key + 'enrichedCommunityGOterms.csv')
this_hyp.community_stats(prefix=key).to_csv(key + '_communityStats.csv')
res = hyStats.compute_all_distances({'mutations': this_hyp})
res.to_csv('panPDXDistances.csv')
nmi = hyStats.compute_all_nmi({'mutations': this_hyp}, g)
nmi.to_csv('panPDXNMI.csv')
def main():
args = parser.parse_args()
if args.fromFile is None:
hyphae = significant_genes(data, 'AML sample', 'Gene', 'LogFoldChange')
else:
hyphae = loadFromFile(args.fromFile)
#now compute graph distances to ascertain fidelity
if args.getDist:
res = hyStats.compute_all_distances(hyphae)
res.to_csv('proteomicdistances.csv')
nmi = hyStats.compute_all_nmi(hyphae, g)
nmi.to_csv('proteomic.csv')
for key, this_hyp in hyphae.items():
this_hyp.node_stats().to_csv(key + '_nodelist.csv')
if args.doEnrich:
if len(this_hyp.forest_enrichment) == 0:
for_e = hyEnrich.go_enrich_forests(this_hyp) #SG, ncbi)
this_hyp.assign_enrichment(for_e, type='forest')
for_e.to_csv(key + 'enrichedForestGoTerms.csv')
this_hyp._to_file(key + '_hypha.pkl')
if len(this_hyp.community_enrichment) == 0:
com_e = hyEnrich.go_enrich_communities(this_hyp)
this_hyp.assign_enrichment(com_e, type='community')
this_hyp._to_file(key + '_hypha.pkl')
com_e.to_csv(key + 'enrichedCommunityGOterms.csv')
##next: compare enrichment between patients mapped to communities
this_hyp.community_stats(prefix=key).to_csv(key +
'_communityStats.csv')
this_hyp.forest_stats().to_csv(key + '_TreeStats.csv')
def main():
args = parser.parse_args()
gfile = args.graph
g = pickle.load(open(gfile, 'rb'))#hyp.make_graph_from_dict(gfile)
if args.fromFile is None:
hyphae = build_hyphae_from_data(args.qt, g, args.sample)
else:
hyphae = loadFromFile(args.fromFile)
#now compute graph distances to ascertain fidelity
if args.getDist:
res = hyStats.compute_all_distances(hyphae)
res.to_csv('panCancerDistances.csv')
nmi = hyStats.compute_all_nmi(hyphae, g)
nmi.to_csv('panCancerNMI.csv')
for key, this_hyp in hyphae.items():
this_hyp.node_stats().to_csv(key+'_nodelist.csv')
if args.doEnrich:
if len(this_hyp.community_enrichment) == 0:
com_e = hyEnrich.go_enrich_communities(this_hyp)
this_hyp.assign_enrichment(com_e, type='community')
this_hyp._to_file(key+'_hypha.pkl')
com_e.to_csv(key+'enrichedCommunityGOterms.csv')
##next: compare enrichment between patients mapped to communities
this_hyp.forest_stats().to_csv(key+'_TreeStats.csv')
this_hyp.community_stats(prefix=key).to_csv(key+'_communityStats.csv')
def main():
args = parser.parse_args()
qval = args.quant
##first we run a helper function to make sure we have all cptac data
fdict = pdat.cptacData()
##first get proteomics measurements
allDat = pdat.getCancerData(fdict, qval, byType=True)
##make srue this file is built!
g = pickle.load(open('../odata/igraphPPI.pkl', 'rb'))
beta = .5
#build hyphal network of network communities
phyph = pickle.load(open(args.hyph, 'rb'))
hDict = {'panCan': phyph}
for ct, dat in allDat.items():
hDict[ct] = hyphalNetwork(dat, g, beta)
nmi = hyStats.compute_all_nmi(hDict, g)
nmi.to_csv(args.refName + '_nmi.csv')
def main():
gfile = '../../data/igraphPPI.pkl'
g = pickle.load(open(gfile, 'rb')) #hyp.make_graph_from_dict(gfile)
args = parser.parse_args()
beta = 0.5
proteomics_dictionary = significant_prots(data, 'AML sample', 'Gene',
'LogFoldChange')
gene_dictionary = tumor_genes(data, 'AML sample', 'Gene', 'Tumor VAF')
if args.fromFile is None:
hyphae = dict()
hyphae['mutations'] = hyphalNetwork(gene_dictionary, g.copy(), beta)
hyphae['proteomics'] = hyphalNetwork(proteomics_dictionary, g.copy(),
beta)
for key, this_hyp in hyphae.items():
this_hyp._to_file(key + '_amlPatientData_hypha.pkl')
else:
hyphae = loadFromFile(args.fromFile)
#now compute graph distances to ascertain fidelity
if args.getDist:
res = hyStats.compute_all_distances(hyphae)
res.to_csv('amlNetworkdistances.csv')
tab = table.build_table("AML Network Distances", 'syn22128879', res)
syn.store(tab)
nmi = hyStats.compute_all_nmi(hyphae, g)
nmi.to_csv('amlNMI.csv')
syn.store(File('amlNMI.csv', parent='syn22269875'))
#store distances
for key, this_hyp in hyphae.items():
node_stats = this_hyp.node_stats()
node_stats.to_csv(key + '_nodelist.csv')
tab = table.build_table("AML Network Nodes", 'syn22128879', node_stats)
syn.store(tab)
if args.doEnrich:
if len(this_hyp.forest_enrichment) == 0:
for_e = hyEnrich.go_enrich_forests(this_hyp) #SG, ncbi)
this_hyp.assign_enrichment(for_e, type='forest')
for_e.to_csv(key + 'enrichedForestGoTerms.csv')
syn.store(
File(key + 'enrichedForestGoTerms.csv',
parent='syn22269875'))
this_hyp._to_file(key + '_amlPatientData_hypha.pkl')
if len(this_hyp.community_enrichment) == 0:
com_e = hyEnrich.go_enrich_communities(this_hyp)
this_hyp.assign_enrichment(com_e, type='community')
com_e.to_csv(key + 'enrichedCommunityGOterms.csv')
syn.store(
File(key + 'enrichedCommunityGOterms.csv',
parent='syn22269875'))
this_hyp._to_file(key + '_amlPatientData_hypha.pkl')
##next: compare enrichment between patients mapped to communities
this_hyp.community_stats(prefix=key).to_csv(key +
'_communityStats.csv')
this_hyp.forest_stats().to_csv(key + '_TreeStats.csv')
for files in [
key + '_amlPatientData_hypha.pkl', key + '_communityStats.csv',
key + '_TreeStats.csv'
]:
syn.store(File(files, parent='syn22269875'))