try:

float(s)

return True

except ValueError:

pfile = open(file1, 'r')

# Gene ID - Profile

# Gene ID include gene symbol or entrez gene ID

pDict = {}

for l in pfile.readlines() :

tp = l.split()

pDict[tp[0]]=map(float, tp[1:])

# keggDict : Kegg ID - Profile

# kgDict : Kegg ID - Gene ID

temp = {}

keggFile = open(KEGGgeneFilePath, 'r')

nullP = []

for i in range(numTP) :

nullP.append(0)

for l in keggFile.readlines() :

tp = l.split("\t")

tp2 = tp[1].split(";")

tp3 = tp2[0].split(", ")

species = tp[0].split(':')[0]

ent = tp[0].split(':')[1]

keggID = tp[0]

firstName = ''

tp3.append(ent)

# If keggID have gene expression

for name in tp3 :

if name==tp3[0] :

firstName = name

if name not in pDict :

continue

keggDict[keggID]=pDict[name]

if name==ent :

kgDict[keggID]=firstName

else :

kgDict[keggID]=name

break

# store gene ID is ith of keggID

for i in range(len(tp3)) :

if tp3[i] not in temp :

temp[tp3[i]]=[]

temp[tp3[i]].append([i, keggID])

# If keggID doesn't have gene expression

if keggID not in keggDict :

keggDict[keggID]=nullP

if firstName=='' :

kgDict[keggID]='None'

else :

kgDict[keggID]=firstName

# if a geneID is match to several keggID,

# choose the keggID with minimum i

for g,list in temp.iteritems() :

min = 10000

for l in list :

if l[0]<min :

gkDict[g]=l[1]

min = l[0]

if np.count_nonzero(p1)==0 or np.count_nonzero(p2)==0 or (type!=1 and type!=-1) :

return False

delay = -1

p1 = np.array(p1)

p2 = np.array(p2)

re = correlate(p1, p2)

if (type==1) :

corr = max(re)

else :

corr = min(re)

if corr==0 :

return False

tot = len(p1)-1

temp = []

for i in range(len(re)) :

if (re[i]==corr) :

temp.append(tot-i)

if min(temp)>=0 : # delay is positive

delay = min([x for x in temp if x>=0])

else : # delay is negative

delay = max([x for x in temp if x<0])

netlist_e = [] # set of entry for each path [ set(0,1,2), set(0,1,2), set(2,3,4) ... ]

netlist_eg = [] # connected components for each path [ (0!g1->1!g2->2!g3), (0!g4->1!g5->2!g6), ... ]

netlist_cc = []

net_e = nx.DiGraph()

net_eg = nx.DiGraph()

for (a,b) in network.edges() :

type = network[a][b]['type']

e1 = entry[a]

e2 = entry[b]

for g1 in e1 :

for g2 in e2 :

if g1 not in keggDict or g2 not in keggDict :

continue

if checkProfile(keggDict[g1], keggDict[g2], type, maxDelay) :

net_e.add_edge(a, b)

net_eg.add_edge(a+'!'+g1, b+'!'+g2)

comps = nx.connected_components(net_eg.to_undirected())

temp = []

for c in comps :

real_c = map(lambda x: x.split('!')[1], c)

real_ce = set(map(lambda x: x.split('!')[0], c))

# Check if the same gene set w/ different entry name exists

if real_c not in temp and ((not skip) or len(real_ce)>2):

temp.append(real_c)

netlist_e.append(set(map(lambda x: x.split('!')[0], c)))

netlist_eg.append(net_eg.subgraph(c))

for eg in netlist_eg :

cc_e = {}

tempcc = 0

for (a, b) in eg.edges() :

e1 = a.split('!')[0]

e2 = b.split('!')[0]

g1 = a.split('!')[1]

g2 = b.split('!')[1]

t1 = np.array(keggDict[g1])

t2 = np.array(keggDict[g2])

if network[e1][e2]['type']==1 :

tempcc = abs(max(correlate(t1, t2)))

else :

tempcc = abs(min(correlate(t1, t2)))

if (e1, e2) not in cc_e :

cc_e[(e1, e2)]=[]

cc_e[(e1, e2)].append(tempcc)

cc = 0

for e, cclist in cc_e.iteritems() :

cc+= sum(cclist)/float(len(cclist))

netlist_cc.append(cc)

labeled = {}

labeled_path = {}

pathDict = {} # [pathway] - pathlist, genepathlist, firstIDs

pathPval = {}

for pathway,net in networks.iteritems() :

print ".. analyzing "+pathway

pathPval[pathway] = []

ent = entries[pathway]

net_e, net_eg, net_cc = checkPathway(net, ent, keggDict, maxDelay, True)

pathDict[pathway] = (net_e, net_eg)

if len(net_e)==0 :

continue

# maximum number of node = number of entries in the pathway

bin = np.array(range(len(ent)+1))

bin2 = np.array(range(len(ent)*numTP))

prob = np.zeros(len(ent)+1)

prob2 = np.zeros(len(ent)*numTP)

repeat = 1000

#randDict = { key: keggDict[key] for key in DEGs[pathway][1] }

randDict = copy.copy(keggDict)

temp = []

temp2 = []

temp3 = []

for j in range(repeat) :

values = randDict.values()

random.shuffle(values)

it = iter(values)

for key in randDict :

randDict[key]=next(it)

temp_e, temp_eg, temp_cc = checkPathway(net, ent, randDict, maxDelay, False)

for c in temp_e :

prob[len(c)]+=1

temp.append(len(c))

for c in temp_cc :

if profile =='DEG' :

prob2[int(round(c))]+=1 # for poisson

temp2.append(c) # for hist

# Fitting the Poisson dist (node)

#prob = prob / np.sum(prob)

#param, cov = curve_fit(poisson_new, bin, prob)

# (cc)

if profile=='DEG' :

if (np.sum(prob2)==0) : # no subpath found in permutation

for i in range(len(net_e)) :

pathPval[pathway].append(0.0)

continue

prob2 = prob2 / np.sum(prob2)

param2, cov2 = curve_fit(poisson_new, bin2, prob2)

for i in range(len(net_e)) :

if profile == 'fold' :

pval = len([x for x in temp2 if x>net_cc[i]])/float(len(temp2)) # cc and histogram

else :

pval = scipy.stats.poisson.sf(round(net_cc[i]), param2[0]) # cc and poisson

pathPval[pathway].append(pval)

print 'P-value:', pval

# Fixed

#if pval < threshold and DEGPval[pathway] < threshold :

if pval < threshold :

source = set([x for x in net_eg[i].nodes() if net_eg[i].in_degree(x)==0])

targets = set(map(lambda x: x.split('!')[1], net_eg[i].nodes()))

print 'source '+str(len(source))

for gene in source :

realgene = gene.split('!')[1]

if realgene not in labeled :

labeled[realgene] = set()

if pathway not in labeled_path :

labeled_path[pathway] = set()

labeled[realgene]|=targets

labeled_path[pathway].add(kgDict[realgene])

networks = {}

entries = {}

DEGs = {} # pathName -> [DEGSet, geneSet]

for root,dirs,files in os.walk(xmlPath):

for file in files:

pathName = file.split(".")[0]

networks[pathName]=nx.DiGraph()

xmlfile = open(os.path.join(xmlPath, file), "r")

xmldata = xmlfile.read()

dom = parseString(xmldata)

xmlfile.close()

geneSet = set()

entries[pathName] = {}

entrydic = entries[pathName]

DEGs[pathName] = [set(),set()]

domentry = dom.getElementsByTagName("entry")

for e in domentry :

if (e.attributes.getNamedItem("type").nodeValue == 'gene') :

id = e.attributes.getNamedItem("id").nodeValue

id = str(id)

genes = e.attributes.getNamedItem("name").nodeValue

genes = str(genes)

genelist = genes.split()

entrydic[id]=set()

for g in genelist :

DEGs[pathName][1].add(g)

if g in keggDict and np.count_nonzero(keggDict[g])!=0 :

DEGs[pathName][0].add(g)

entrydic[id].add(g)

elif (e.attributes.getNamedItem("type").nodeValue == 'group') :

id = e.attributes.getNamedItem("id").nodeValue

id = str(id)

comps = e.getElementsByTagName("component")

entrydic[id]=set()

for c in comps :

geneId =c.attributes.getNamedItem("id").nodeValue

for g in entrydic[geneId] :

entrydic[id].add(g)

relations = dom.getElementsByTagName("relation")

for r in relations :

subs = r.getElementsByTagName("subtype")

ent1 = r.attributes.getNamedItem("entry1").nodeValue

ent2 = r.attributes.getNamedItem("entry2").nodeValue

if (ent1==ent2) or (ent1 not in entrydic) or (ent2 not in entrydic) :

continue

if (not (subs==[])) :

for s in subs :

type = s.attributes.getNamedItem("name").nodeValue

j = 0

if (type=="activation" or type=="expression" or type=='indirect effect') and len(entrydic[ent1])!=0 and len(entrydic[ent2])!=0:

j=1

elif (type=="inhibition" or type=="repression") and len(entrydic[ent1])!=0 and len(entrydic[ent2])!=0 :

j=-1

elif (type=="compound" or type=="binding/association") and (len(entrydic[ent1])!=0 or len(entrydic[ent2])!=0) :

j=2 # !!!

if j==1 or j==-1 :

networks[pathName].add_edge(ent1, ent2, type=j)

elif j==2 :

networks[pathName].add_edge(ent1, ent2, type=j)

networks[pathName].add_edge(ent2, ent1, type=j)

(M, n, N) = args[0:3]

try:

return scipy.stats.hypergeom.sf(k, *args, **kwds)

except Exception as inst:

if k >= n and type(inst) == IndexError:

return 0 ## or conversely 1 - hypergeom.cdf(k, *args, **kwds)

else:

totDEG = set()

totGene = set()

for path, l in DEGs.iteritems() :

totDEG = totDEG | l[0]

totGene = totGene | l[1]

totDEGN = len(totDEG)

totGeneN = len(totGene)

DEGPval = {}

for path, l in DEGs.iteritems() :

A = len(l[0])

B = totDEGN-A

C = len(l[1])-A

D = totGeneN-totDEGN-C

odds, pval = scipy.stats.fisher_exact([[A,B],[C,D]])

DEGPval[path]=pval

try:

float(s)

return True

except ValueError:

GRNfile = open(file1, 'r')

PPIfile = open(file2, 'r')

negProfit = 0

TFset = set()

for l in GRNfile.readlines() :

tp = l.rstrip().split()

TFset.add(tp[0])

TFset.add(tp[0].replace("-", "").upper())

if species+':'+tp[0] in kgDict :

TFset.add(kgDict[species+':'+tp[0]])

GRNfile.close()

GRNfile = open(file1, 'r')

for l in GRNfile.readlines() :

tp = l.rstrip().split()

TF = tp[0]

gene = tp[1]

# add genes (found in the gkDict) to the network

if TF not in gkDict and (TF.replace("-", "").upper() in gkDict) :

TF = TF.replace("-", "").upper()

# TF & gene should be in kegg

if TF not in gkDict or gene not in gkDict :

continue

match = checkProfile(keggDict[gkDict[TF]], keggDict[gkDict[gene]], 1, maxDelay)

match2 = checkProfile(keggDict[gkDict[TF]], keggDict[gkDict[gene]], -1, maxDelay)

if match :

type = 1

elif match2 :

type = -1

else :

continue

label = len(labeled[gkDict[TF]]) if gkDict[TF] in labeled else negProfit

TF = kgDict[gkDict[TF]]

network.add_node(TF, label=label, isTF=1)

label = len(labeled[gkDict[gene]]) if gkDict[gene] in labeled else negProfit

gene = kgDict[gkDict[gene]]

isTF = 1 if gene in TFset else 0

network.add_node(gene, label=label, isTF=isTF)

network.add_edge(TF, gene, type=type, weight=1000)

for l in PPIfile.readlines() :

tp = l.rstrip().split()

g1 = tp[0]

g2 = tp[1]

if g1 not in gkDict or g2 not in gkDict :

continue

# Change weight only for PPI

rType = -1 if tp[3]=='inhibition' else 1

match = checkProfile(keggDict[gkDict[g1]], keggDict[gkDict[g2]], rType, maxDelay)

weight = int(tp[5]) if len(tp)>=6 and is_number(tp[5]) else int(tp[4])

if not match :

continue

#weight = weight*0.5

tp1 = g1

tp2 = g2

# Change gene name for consistency with gene labels from expression data

label = len(labeled[gkDict[g1]]) if gkDict[g1] in labeled else negProfit

g1 = kgDict[gkDict[g1]]

isTF = 1 if g1 in TFset else 0

network.add_node(g1, label=label, isTF=isTF)

label = len(labeled[gkDict[g2]]) if gkDict[g2] in labeled else negProfit

g2 = kgDict[gkDict[g2]]

isTF = 1 if g2 in TFset else 0

network.add_node(g2, label=label, isTF=isTF)

if tp[3]=='activation' :

network.add_edge(g1, g2, type=1, weight=weight)

elif tp[3]=='inhibition' :

network.add_edge(g1, g2, type=-1, weight=weight)

else :

network.add_edge(g1, g2, type=1, weight=weight)

network.add_edge(g2, g1, type=1, weight=weight)

# edge type : -1/1 inhibition/activation, 0 TF-target(TF/gene), 2 else PPI

sList = g.successors(n)

infSet = set([n])

profit = g.node[n]['label']

if len(sList)==0 :

return infSet, profit

for s in sList :

if s in visited :

continue

visited.add(s)

(a, b) = influenceByNode(s, g, visited)

infSet = infSet | a

profit += b

p = keggDict[gkDict[n]]

pSet = set()

nSet = set()

for i in range(len(p)) :

if p[i]==1 :

pSet.add(i)

elif p[i]==-1 :

nSet.add(i)

p = keggDict[gkDict[n]]

if type==1 :

prevP = pair[0]

prevN = pair[1]

elif type==-1 :

prevP = pair[1]

prevN = pair[0]

pSet = set()

nSet = set()

for ind in prevP :

for i in range(ind, min(ind+maxDelay+1,len(p))):

if p[i]==1 :

pSet.add(i)

for ind in prevN :

for i in range(ind, min(ind+maxDelay+1,len(p))):

if p[i]==-1 :

nSet.add(i)

visited = set()

infNet = nx.DiGraph()

pnDict = {}

profit = 0

stack = []

stack.append(n)

pnDict[n] = startInd(n)

while len(stack)>0:

v = stack.pop()

if v not in visited :

visited.add(v)

prevP, prevN = pnDict[v]

for s in g.successors(v) :

p, n = nextInd(s, (prevP, prevN), g[v][s]['type'], maxDelay)

if len(p)+len(n) > 0 :

infNet.add_edge(v, s)

profit += g.node[s]['label']

stack.append(s)

pnDict[s] = (p, n)

infNo = {}

TFpath = {}

TFSet = set([x for x in network.nodes() if network.node[x]['isTF']==1])

repeat = 100

seedSet = set()

elseSet = TFSet

newSeed = ''

for j in range(k) :

print 'TF '+str(j)

for n in elseSet :

infNo[n]=0

for i in range(repeat) :

# Produce G'

g = network.copy()

for (a,b) in network.edges() :

if random.randint(1,1000)>g[a][b]['weight'] :

g.remove_edge(a, b)

# Influence set by seed

seedInfSet = set()

for TF in seedSet :

(infNet, profit) = infByNode(TF, g, maxDelay)

infSet = infNet.nodes()

seedInfSet = seedInfSet | set(infSet)

# If a TF is not influenced by current seeds then calculate profit

for n in elseSet-seedInfSet :

(infNet, profit) = infByNode(n, g, maxDelay)

infSet = infNet.nodes()

if len(infSet)>0 :

infNo[n] += float(profit)/len(infSet)

for n in elseSet :

infNo[n]=float(infNo[n])/repeat

newSeed = max(infNo.iterkeys(), key=(lambda x : infNo[x]))

if infNo[newSeed]==0.0 :

continue

infNet, profit = infByNode(newSeed, network, maxDelay)

TFpath[newSeed]=infNet

print newSeed+'\t'+str(infNo[newSeed])

seedSet.add(newSeed)

elseSet.remove(newSeed)

del infNo[newSeed]

url = "http://www.kegg.jp/kegg-bin/show_pathway?map="+pathway+"&multi_query="

for gene in net_eg.nodes() :

url += gene.split('!')[1]+"+red%0D%0A"

temp = []

for i in range(len(profile)) :

color = '#ffffff'

if profile[i]>=1.0 :

color = '#ff0000'

elif profile[i]>=0.8 and profile[i]<1.0:

color = '#ff2a2a'

elif profile[i]>=0.6 and profile[i]<0.8 :

color = '#ff5555'

elif profile[i]>=0.4 and profile[i]<0.6:

color = '#ff7f7f'

elif profile[i]>=0.2 and profile[i]<0.4 :

color = '#ffaaaa'

elif profile[i]>0 and profile[i]<0.2:

color = '#ffd4d4'

elif profile[i]==0 :

color = '#ffffff'

elif profile[i]>-0.2 and profile[i]<0:

color = '#d4d4ff'

elif profile[i]>-0.4 and profile[i]<=-0.2 :

color = '#aaaaff'

elif profile[i]>0.6 and profile[i]<=-0.4:

color = '#7f7fff'

elif profile[i]>0.8 and profile[i]<=-0.6 :

color = '#5555ff'

elif profile[i]>-1.0 and profile[i]<=-0.8:

color = '#2a2aff'

elif profile[i]<=-1.0 :

color = '#0000ff'

temp.append('p'+str(i+1)+':"'+color+'"')

file1 = open(webPath+'/'+pathway+'_'+str(pathNo+1)+'.html', 'w')

file1.write('<head><link href="style.css" rel="stylesheet" />\n')

file1.write('<script src="http://ajax.googleapis.com/ajax/libs/jquery/1/jquery.min.js"></script>\n')

file1.write('<script src="https://cdn.rawgit.com/cytoscape/cytoscape.js/d48294ac/dist/cytoscape.min.js"></script>\n')

file1.write('<script src="https://unpkg.com/dagre@0.7.4/dist/dagre.js"></script>\n<script src="cytoscape-dagre.js"></script>\n')

file1.write('<script src="'+pathway+'_'+str(pathNo+1)+'.js"></script></head>\n')

file1.write('<body><div id="cy"></div></body></html>\n')

file1.close()

file2 = open(webPath+'/'+pathway+'_'+str(pathNo+1)+'.js', 'w')

file2.write('window.addEventListener(\'DOMContentLoaded\', function(){\nvar cy = window.cy = cytoscape({\n container: document.getElementById(\'cy\'),\n boxSelectionEnabled: false, autounselectify: true, layout: { name: \'dagre\' },\n')

file2.write("style: [ { selector: \'node[type=\"single\"]\',\nstyle: {\"width\":\"30px\",\"height\":\"30px\",\"content\":\"data(id)\",\"border-color\":\"data(bc)\",\"border-width\": 3,\"pie-size\":\"100%\",\n")

for i in range(numTP) :

file2.write('"pie-'+str(i+1)+'-background-color": "data(p'+str(i+1)+')",\n')

file2.write('"pie-'+str(i+1)+'-background-size": '+str(100/(numTP)))

if i<numTP-1 :

file2.write(',')

else:

file2.write('}\n},\n')

file2.write('{ selector: \'edge\',\nstyle:{\n"width":"mapData(weight, 0, 1000, 1, 4)","target-arrow-shape":"data(shape)","opacity":1.0,"target-arrow-color":"black","curve-style":"bezier"}\n}],\n')

nodes = set()

edges = set()

TFs = set()

totTFs = set([x for x in network.nodes() if network.node[x]['isTF']==1])

# TF to the first gene

sources = set([x for x in net_eg.nodes() if net_eg.in_degree(x)==0])

for gene in sources :

first = kgDict[gene.split('!')[1]]

minLen = float('Inf')

minSeed = ''

for seed in seedSet :

if first not in TFpath[seed].nodes() :

continue

try:

length = nx.shortest_path_length(TFpath[seed], seed, first)

except:

length = float('Inf')

continue

if length < minLen :

minLen = length

minSeed = seed

if minSeed =='' :

continue

TFs.add(minSeed)

spath = nx.shortest_path(TFpath[minSeed], minSeed, first)

for i in range(len(spath)-1) :

if i==0 : # TF

nodes.add('{ data: { id: "'+spath[i]+'",'+profilejs(keggDict[gkDict[spath[i]]])+', bc: "blue", type: "single" }}')

else : # other

if network.node[spath[i]]['isTF']==1 :

color = 'blue'

else :

color = 'gray'

nodes.add('{ data: { id: "'+spath[i]+'",'+profilejs(keggDict[gkDict[spath[i]]])+', bc: "'+color+'", type: "single" }}')

if i<len(spath)-1 :

shape = 'tee' if network[spath[i]][spath[i+1]]['type']==-1 else 'triangle'

edges.add('{ data: { source: "'+spath[i]+'", target: "'+spath[i+1]+'", weight:'+str(network[spath[i]][spath[i+1]]['weight'])+', shape:"'+shape+'"}}')

# perturbed path

egDict = {}

for (g1, g2) in net_eg.edges() :

eg1 = g1.split('!')

eg2 = g2.split('!')

e1 = eg1[0]

e2 = eg2[0]

g1 = eg1[1]

g2 = eg2[1]

if e1 not in egDict :

egDict[e1] = set()

egDict[e1].add(g1)

if e2 not in egDict :

egDict[e2] = set()

egDict[e2].add(g2)

g1 = kgDict[eg1[1]]

g2 = kgDict[eg2[1]]

shape = 'tee' if knetwork[e1][e2]['type']==-1 else 'triangle'

edges.add('{ data: { source: "'+g1+'", target: "'+g2+'", weight: 1000, shape:"'+shape+'"}}')

for e, gSet in egDict.iteritems() :

# if it has to be a compound node

if len(gSet)>1 :

nodes.add('{data:{id:"'+e+'", type: "compound"}}')

for g in gSet :

nodes.add('{data:{id:"'+kgDict[g]+'",'+profilejs(keggDict[g])+', parent:"'+e+'", bc: "green", type: "single" }}')

# single node

else :

g = gSet.pop()

nodes.add('{data:{id:"'+kgDict[g]+'",'+profilejs(keggDict[g])+', bc: "green", type: "single" }}')

# Print out

file2.write('elements:{ nodes: ['+','.join(list(nodes))+'],\n edges: ['+','.join(list(edges))+']}\n});\n});')

file2.close()

pnameFile = open(KEGGPathFilePath,'r')

pnameDic = {}

for path in pnameFile.readlines() :

tp = path.split("\t")

tp2 = tp[0].split(":")

tp3 = tp[1].split(" - ")

pnameDic[tp2[1]]=tp3[0]

pnameFile.close()

file1 = open(os.path.join(resPath, 'trap_pvalue.html'), 'w')

file1.write('<!DOCTYPE html>\n<html>\n<head><link href="style.css" rel="stylesheet" /></head>\n<body>\n<table>\n')

file1.write('<tr><th>Pathway</th><th class="name">Pathway name</th><th>DEG p-value</th><th>Path</th><th>Path p-value</th><th>Combined p-value</th><th>Link1</th><th>Link2</th><th>TFs</th></tr>\n')

strs = []

for pathway,tri in pathDict.iteritems() :

pathlist = tri[0] # the list of entry path

genelist = tri[1] # the network (ent!gene)

if pathway not in pnameDic :

continue

p = pnameDic[pathway]

if len(p)>30 :

p = p[:30]+'...'

s = ''

row = str(1) if len(pathlist)<2 else str(len(pathlist))

s+='<tr><td rowspan="'+row+'">'+pathway+'</td><td class="name" rowspan="'+row+'">'+p+'</td><td rowspan="'+row+'">'+ "{:5.3f}".format(DEGPval[pathway])+'</td>\n'

# If there is perturbed path(s)

for i in range(len(pathlist)) :

st, comPval = scipy.stats.combine_pvalues([DEGPval[pathway], pathPval[pathway][i]])

if i!=0 :

s+='<tr>'

pstr = "{:5.3f}".format(pathPval[pathway][i])

cpstr = "{:5.3f}".format(comPval)

if pathPval[pathway][i]< threshold :

pstr = '<b>'+pstr+'</b>'

if comPval< threshold :

cpstr = '<b>'+cpstr+'</b>'

s+='<td>Path'+str(i+1)+'</td><td>'+pstr+'</td><td>'+cpstr+'</td>\n'

link1 = keggDiagram(pathway, genelist[i])

link2, TFs = cytoscapejs(resPath, pathway, i, pathlist[i], genelist[i], networks[pathway], network, seedSet, numTP, TFpath)

s+='<td><a href="'+link1+'" target="diagram">KEGG</td>\n<td><a href="'+link2+'" target="cyto">CYTO</a></td><td>'+','.join(list(TFs))+'</td></tr>\n'

# If there is no perturbed path

if len(pathlist)==0 :

for i in range(6) :

if i==2 :

s+='<td>'+"{:5.3f}".format(DEGPval[pathway])+'</td>'

else :

s+='<td></td>'

s+='</tr>\n'

strs.append(s)

strs = sorted(strs, key = lambda x: float(re.findall(r"[0-1]\.[0-9]+", x)[0]))

file1.write('\n'.join(strs))

file1.write('</table></body></html>\n')

file1.close()

file1 = open(webPath+'/network_overview.html', 'w')

file1.write('<head><link href="style.css" rel="stylesheet" />\n')

file1.write('<script src="http://ajax.googleapis.com/ajax/libs/jquery/1/jquery.min.js"></script>\n')

file1.write('<script src="https://cdn.rawgit.com/cytoscape/cytoscape.js/d48294ac/dist/cytoscape.min.js"></script>\n')

file1.write('<script src="https://unpkg.com/dagre@0.7.4/dist/dagre.js"></script>\n<script src="cytoscape-dagre.js"></script>\n')

file1.write('<script src="network_overview.js"></script></head>\n')

file1.write('<body><div id="cy"></div></body></html>\n')

file1.close()

file2 = open(webPath+'/network_overview.js', 'w')

file2.write('window.addEventListener(\'DOMContentLoaded\', function(){\nvar cy = window.cy = cytoscape({\n container: document.getElementById(\'cy\'),\n boxSelectionEnabled: false, autounselectify: true, layout: { name: \'dagre\' },\n')

file2.write("style: [ { selector: \'node[type = \"single\"]',\nstyle: {\"width\":\"30px\",\"height\":\"30px\",\"content\":\"data(id)\",\"border-color\":\"data(bc)\",\"border-width\": 3,\"pie-size\":\"100%\",\n")

for i in range(numTP) :

file2.write('"pie-'+str(i+1)+'-background-color": "data(p'+str(i+1)+')",\n')

file2.write('"pie-'+str(i+1)+'-background-size": '+str(100/(numTP)))

if i<numTP-1 :

file2.write(',')

else:

file2.write('}\n},\n')

file2.write('{ selector: \'edge\',\nstyle:{\n"width":"mapData(weight, 0, 1000, 1, 4)","target-arrow-shape":"data(shape)","opacity":1.0,"target-arrow-color":"black","curve-style":"bezier"}\n}],\n')

tempNet = nx.DiGraph()

nodes = set()

edges = set()

# TF-Pathway

for seed, infNet in TFpath.iteritems() :

targets = set([x for x in infNet.nodes() if network.node[x]['label']>0])

for t in targets :

spath = nx.shortest_path(infNet, seed, t)

for i in range(len(spath)) :

color = 'gray'

if network.node[spath[i]]['isTF']==1 :

color = 'blue'

if network.node[spath[i]]['label']>0 :

color = 'green'

nodes.add('{ data: { id: "'+spath[i]+'",'+profilejs(keggDict[gkDict[spath[i]]])+', bc: "'+color+'", type: "single" }}')

if i<len(spath)-1 :

shape = 'tee' if network[spath[i]][spath[i+1]]['type']==-1 else 'triangle'

edges.add('{ data: { source: "'+spath[i]+'", target: "'+spath[i+1]+'", weight:'+str(network[spath[i]][spath[i+1]]['weight'])+', shape:"'+shape+'"}}')

# Perturbed path

parent = 1

for pathway, tri in pathDict.iteritems() :

if not DEGPval[pathway] < threshold :

continue

knetwork = networks[pathway]

pathlist = tri[0] # the list of entry path

genelist = tri[1] # the network (ent!gene)

for i in range(len(pathlist)) :

if not pathPval[pathway][i] < threshold :

continue

net_eg = genelist[i]

egDict = {}

for (g1, g2) in net_eg.edges() :

eg1 = g1.split('!')

eg2 = g2.split('!')

e1 = eg1[0]

e2 = eg2[0]

g1 = eg1[1]

g2 = eg2[1]

if e1 not in egDict :

egDict[e1] = set()

egDict[e1].add(g1)

if e2 not in egDict :

egDict[e2] = set()

egDict[e2].add(g2)

g1 = kgDict[eg1[1]]

g2 = kgDict[eg2[1]]

shape = 'tee' if knetwork[e1][e2]['type']==-1 else 'triangle'

edges.add('{ data: { source: "'+g1+'", target: "'+g2+'", weight: 1000, shape:"'+shape+'"}}')

for e, gSet in egDict.iteritems() :

# if it has to be a compound node

if len(gSet)>1 :

nodes.add('{data:{id:"'+str(parent)+'", type: "compound"}}')

for g in gSet :

nodes.add('{data:{id:"'+kgDict[g]+'",'+profilejs(keggDict[g])+', parent:"'+str(parent)+'", bc: "green", type: "single" }}')

parent +=1

# single node

else :

g = gSet.pop()

nodes.add('{data:{id:"'+kgDict[g]+'",'+profilejs(keggDict[g])+', bc: "green", type: "single" }}')

# Print out

file2.write('elements:{ nodes: ['+','.join(list(nodes))+'],\n edges: ['+','.join(list(edges))+']}\n});\n});')

empty = open(webPath+'/empty.html','w')

index = open(webPath+'/index.html','w')

css = open(webPath+'/style.css','w')

index.write('<!DOCTYPE html>\n<html><head><title>TRAP results</title><link href="style.css" rel="stylesheet" /></head>\n<frameset cols="850px,*">\n<frame src="trap_pvalue.html" name="left" id="left" frameborder="0" />\n<frameset rows="50%, 50%">\n<frame src="empty.html" name="diagram" id="diagram" frameborder="0" />\n <frame src="empty.html" name="cyto" id="cyto" frameborder="0" />\n</frameset>\n</frameset>\n<body></body>\n</html>\n')

css.write('body { font: 14px helvetica neue, helvetica, arial, sans-serif; }\nth, td { border: 1px black; width: 100px; text-align:center; }\n .name { width: 300px;}\n td.name { text-align: left; }\n#cy { height: 100%; width: 100%; position: absolute; left: 0; top: 0; }\n')

empty.close()

index.close()