def dose_response_caller(gct_file,verbose=True):
'''
Counts the number of dose-responsive genes based on template matching
also examines the S-C correlation to determine if a perturbation is dose
responsive
'''
dp = doseClass.DosePlate()
dp.add_from_gct(gct_file)
dp.examine_doses_tested(verbose=verbose)
dp.match_template(verbose=verbose)
# dp.permutation_template()
ssAll = dp.gct.get_column_meta('distil_ss')
ccAll = dp.gct.get_column_meta('distil_cc_q75')
#matric what is the dose corr
nMatchDict = {}
for pert in dp.perts_at_dose:
IndsPert = dp.doseIndDict[pert]
ssVals = [float(ssAll[i]) for i in IndsPert]
ccVals = [float(ccAll[i]) for i in IndsPert]
corr, p = stats.pearsonr(ccVals,ssVals)
if p < .2:
print pert + ' S-C corr ' + str(corr)
#find number of probes that match a template pass FDR
templates = dp.templateMatchInd[pert].keys()
probeTempMatches = []
for temp in templates:
probeTempMatches.extend(dp.templateMatchInd[pert][temp])
probeMatch = set(probeTempMatches)
print pert + ' number of template matches ' + str(len(probeMatch))
### DR caller
if len(probeMatch) > 10 or p < .1:
print pert + ' is dose responsive'
nMatchDict[pert] = len(probeMatch)
return nMatchDict
ominWithContext.append(geneID)
### in which cell lines do we have more than 500 CGS sigs
cellFull = [q['cell_id'] for q in CGSall]
cgsCellsLong = list(set(cellFull))
cgsCells = []
for cell in cgsCellsLong:
if cellFull.count(cell) > 500:
cgsCells.append(cell)
### get HOG brds
file1 = '/xchip/obelix/pod/brew/pc/HOG001_A549_24H/by_pert_id_pert_dose/HOG001_A549_24H_COMPZ.MODZ_SCORE_LM_n288x978.gctx'
file2 = '/xchip/obelix/pod/brew/pc/HOG002_A549_24H/by_pert_id_pert_dose/HOG002_A549_24H_COMPZ.MODZ_SCORE_LM_n288x978.gctx'
hogPerts = []
#plate 1
dp = doseClass.DosePlate()
dp.add_from_gct(file1)
dp.examine_doses_tested()
hogPerts.extend(dp.perts_at_dose)
#plaste 2
dp = doseClass.DosePlate()
dp.add_from_gct(file2)
dp.examine_doses_tested()
hogPerts.extend(dp.perts_at_dose)
hogPertsShort = [x[:13] for x in hogPerts]
### pull sigs of hog perts from mongo
# for pert in hogPertsShort:
# for cell in cgsCells:
# CM = mutil.CMapMongo()
# cpSigs = CM.find({'pert_id':pert,'pert_type':'trt_cp'},{'sig_id':True,'cell_id':True,'pert_iname':True,'pert_id':True})
# refControl = 'pc' #use pc vs vc controled data
# gctfile = glob.glob('%s/%s/%s_%s_%s/by_pert_id_pert_dose/%s_%s_%s_COMPZ.MODZ_SCORE_LM_*.gctx' % (data_dir,refControl,platePrefix,cell,timeP,platePrefix,cell,timeP))
# gctfile = gctfile[0]
# work_dir = '/xchip/cogs/projects/HOG/scratch/%s_%s_%s' % (cell,timeP,refControl)
# if not os.path.exists(work_dir):
# os.mkdir(work_dir)
# db = gct.GCT() #make a gct object
# db.read(gctfile)
###merged gct
gctfile = '/xchip/cogs/projects/HOG/data/brew_by_pert_id_pert_dose/pc/HOG_merged_MCF7_24H/merged_COMPZ.MODZ_SCORE_LM_n862x978.gctx'
db = gct.GCT() #make a gct object
db.read(gctfile)
do = doseClass.DosePlate()
do.add_from_gct()
pertIDs = db.get_column_meta('pert_id')
pertDescs = db.get_column_meta('pert_desc')
### parp group
# XAV939 BRD-K12762134-001-02-1
# veliparib BRD-K87142802-001-02-7
# olaparib BRD-K02113016
# 3-amino-benzamide BRD-K08703257-001-06-3
# IWR-1-endo BRD-K61314889-001-01-0
# NU-1025 BRD-K48692744-001-01-2
# DR2313 BRD-K94920105-001-01-3
# phenanthridinone BRD-K11163873-001-02-8
# PJ34 BRD-K11853856-003-01-3