def formatAndSubmitSplicingEventsToSashimiPlot(filename, bamdir,
splicing_events,
sample_group_db, groups,
expandedSearch):
### Begin exporting parameters and events for SashimiPlot visualization
firstLine = True
setting = unique.filepath("Config/sashimi_plot_settings.txt")
psi_parent_dir = findParentDir(filename)
if 'PSI' not in filename:
index_dir = string.split(
psi_parent_dir, 'ExpressionInput'
)[0] + "AltResults/AlternativeOutput/sashimi_index/"
else:
index_dir = psi_parent_dir + "sashimi_index/"
spliced_junctions = [
] ### Alternatively, compare to just one of the junctions
for splicing_event in splicing_events:
try:
j1, j2 = string.split(splicing_event, ' ')
spliced_junctions.append(j1)
spliced_junctions.append(j2)
except Exception:
spliced_junctions.append(
splicing_event) ### single gene ID or junction
if 'PSI' not in filename:
splicing_events_db = {}
for event in splicing_events:
event = string.replace(event, ':', '__')
if ' ' in event:
event = string.split(event, ' ')[-1]
gene = string.split(event, "__")[0]
try:
splicing_events_db[gene].append(event)
except Exception:
splicing_events_db[gene] = [event]
splicing_events = splicing_events_db
import collections
analyzed_junctions = []
processed_events = []
#restrictToTheseGroups=['WT','R636S_Het','R636S_Homo'] #Meg HSCP-1 , Myelocyte Mono
restrictToTheseGroups = None
for line in open(filename, 'rU').xreadlines():
line = cleanUpLine(line)
t = string.split(line, '\t')
if firstLine:
if 'PSI' in filename:
sampleIndexBegin = 11
sample_headers = t[sampleIndexBegin:]
else:
sampleIndexBegin = 1
sample_headers = t[sampleIndexBegin:]
if '.bed' not in sample_headers[
0]: ### Add .bed if removed manually
sample_headers = map(lambda s: s + '.bed', sample_headers)
index = 0
sample_group_index = {}
for s in sample_headers:
group = sample_group_db[s]
sample_group_index[index] = group
try:
sampleReadDepth[index] = count_sum_array_db[s]
except Exception:
sampleReadDepth[index] = count_sum_array_db[s]
index += 1
firstLine = False
else:
if 'PSI' in filename:
splicing_event = val = t[2] + ' ' + t[3]
j1 = t[2]
j2 = t[3]
if t[2] in analyzed_junctions and t[3] in analyzed_junctions:
continue
else:
splicing_event = t[0] ### The gene ID
j1 = t[0]
j2 = t[0]
if ":U" in splicing_event or "-U" in splicing_event:
continue
else:
### First check to see if the full splicing event matches the entry
### If not (and not a PSI regulation hits list), look for an individual junction match
if splicing_event in splicing_events or (
expandedSearch and
(j1 in spliced_junctions or j2 in spliced_junctions)):
if splicing_event in processed_events: continue
if j2 in processed_events: continue
if j1 in processed_events: continue
processed_events.append(splicing_event)
processed_events.append(j1)
processed_events.append(j2)
#print processed_events, splicing_event
if 'PSI' in filename:
geneID = string.split(t[2], ':')[0]
symbol = t[0]
analyzed_junctions.append(t[2])
analyzed_junctions.append(t[3])
else: ### For exp.dataset-steady-state.txt files
geneID = splicing_event
events = splicing_events[geneID]
index = 0
import collections
initial_group_psi_values = {}
try:
group_psi_values = collections.OrderedDict()
except Exception:
try:
import ordereddict
group_psi_values = ordereddict.OrderedDict()
except Exception:
group_psi_values = {}
for i in t[
sampleIndexBegin:]: ### Value PSI range in the input file
try:
group = sample_group_index[index]
except Exception:
group = None
try:
try:
initial_group_psi_values[group].append(
[float(i), index])
except Exception:
initial_group_psi_values[group] = [[
float(i), index
]]
except Exception:
#print traceback.format_exc();sys.exit()
pass ### Ignore the NULL values
index += 1
if restrictToTheseGroups != None: ### Exclude unwanted groups
initial_group_psi_values2 = {}
groups2 = collections.OrderedDict()
for group in groups:
if group in initial_group_psi_values:
if group in restrictToTheseGroups:
initial_group_psi_values2[
group] = initial_group_psi_values[
group]
groups2[group] = []
initial_group_psi_values = initial_group_psi_values2
groups = groups2
### limit the number of events reported and sort based on the PSI values in each group
if 'None' in groups and len(groups) == 1:
initial_group_psi_values['None'].sort()
group_size = len(initial_group_psi_values['None']) / 2
filtered_group_index1 = map(
lambda x: x[1],
initial_group_psi_values['None'][:group_size])
filtered_group_index2 = map(
lambda x: x[1],
initial_group_psi_values['None'][group_size:])
group_psi_values['low'] = filtered_group_index1
group_psi_values['high'] = filtered_group_index2
else:
gn = 0
for group in groups:
#print group
gn += 1
#if gn>4: break
if group in initial_group_psi_values:
initial_group_psi_values[group].sort()
if len(groups) > 7:
filtered_group_indexes = map(
lambda x: x[1],
initial_group_psi_values[group][:1])
elif len(groups) > 5:
filtered_group_indexes = map(
lambda x: x[1],
initial_group_psi_values[group][:2])
elif len(groups) > 3:
filtered_group_indexes = map(
lambda x: x[1],
initial_group_psi_values[group][:4])
else:
filtered_group_indexes = map(
lambda x: x[1],
initial_group_psi_values[group][:5])
group_psi_values[
group] = filtered_group_indexes
try:
update_plot_settings(bamdir, group_psi_values,
sample_headers)
except Exception:
print 'Cannot update the settings file. Likely permissions issue.'
try:
reordered = reorderEvents([t[2] + ' ' + t[3]])
reordered = string.split(reordered[0], ' ')
except Exception:
reordered = [t[2] + ' ' + t[3]]
reordered = string.split(reordered[0], ' ')
#print reordered
if 'PSI' in filename:
try:
formatted_splice_event = string.replace(
reordered[1], ':', '__')
except Exception:
pass
### Submit the query
try:
ssp.plot_event(formatted_splice_event, index_dir,
setting, outputdir)
success = True
except Exception:
success = False
#print traceback.format_exc()
else:
for event in events:
try:
ssp.plot_event(event, index_dir, setting,
outputdir)
#print 'success' #formatted_splice_event='ENSMUSG00000000355__E4.1-E5.1'
except Exception: ### If it fails, output the gene-level plot
try:
ssp.plot_event(geneID, index_dir, setting,
outputdir)
success = True
except Exception:
success = False
#print traceback.format_exc()
"""
### Second attempt
if 'PSI' in filename and success==False: ### Only relevant when parsing the junction pairs but not genes
try: formatted_splice_event=string.replace(reordered[0],':','__')
except Exception: pass
try: ssp.plot_event(formatted_splice_event,index_dir,setting,outputdir); # print 'success'
except Exception: pass
"""
return processed_events
def formatAndSubmitSplicingEventsToSashimiPlot(
filename, bamdir, splicing_events, sample_group_db, groups, expandedSearch
):
### Begin exporting parameters and events for SashimiPlot visualization
firstLine = True
setting = unique.filepath("Config/sashimi_plot_settings.txt")
psi_parent_dir = findParentDir(filename)
if "PSI" not in filename:
index_dir = string.split(psi_parent_dir, "ExpressionInput")[0] + "AltResults/AlternativeOutput/sashimi_index/"
else:
index_dir = psi_parent_dir + "sashimi_index/"
spliced_junctions = [] ### Alternatively, compare to just one of the junctions
for splicing_event in splicing_events:
try:
j1, j2 = string.split(splicing_event, " ")
spliced_junctions.append(j1)
spliced_junctions.append(j2)
except Exception:
spliced_junctions.append(splicing_event) ### single gene ID or junction
if "PSI" not in filename:
splicing_events_db = {}
for event in splicing_events:
event = string.replace(event, ":", "__")
if " " in event:
event = string.split(event, " ")[-1]
gene = string.split(event, "__")[0]
try:
splicing_events_db[gene].append(event)
except Exception:
splicing_events_db[gene] = [event]
splicing_events = splicing_events_db
import collections
analyzed_junctions = []
processed_events = []
for line in open(filename, "rU").xreadlines():
line = cleanUpLine(line)
t = string.split(line, "\t")
if firstLine:
if "PSI" in filename:
sampleIndexBegin = 11
sample_headers = t[sampleIndexBegin:]
else:
sampleIndexBegin = 1
sample_headers = t[sampleIndexBegin:]
if ".bed" not in sample_headers[0]: ### Add .bed if removed manually
sample_headers = map(lambda s: s + ".bed", sample_headers)
index = 0
sample_group_index = {}
for s in sample_headers:
group = sample_group_db[s]
sample_group_index[index] = group
try:
sampleReadDepth[index] = count_sum_array_db[s]
except Exception:
sampleReadDepth[index] = count_sum_array_db[s]
index += 1
firstLine = False
else:
if "PSI" in filename:
splicing_event = val = t[2] + " " + t[3]
j1 = t[2]
j2 = t[3]
if t[2] in analyzed_junctions and t[3] in analyzed_junctions:
continue
else:
splicing_event = t[0] ### The gene ID
j1 = t[0]
j2 = t[0]
if ":U" in splicing_event or "-U" in splicing_event:
continue
else:
### First check to see if the full splicing event matches the entry
### If not (and not a PSI regulation hits list), look for an individual junction match
if splicing_event in splicing_events or (
expandedSearch and (j1 in spliced_junctions or j2 in spliced_junctions)
):
if splicing_event in processed_events:
continue
if j2 in processed_events:
continue
if j1 in processed_events:
continue
processed_events.append(splicing_event)
processed_events.append(j1)
processed_events.append(j2)
# print processed_events, splicing_event
if "PSI" in filename:
geneID = string.split(t[2], ":")[0]
symbol = t[0]
analyzed_junctions.append(t[2])
analyzed_junctions.append(t[3])
else: ### For exp.dataset-steady-state.txt files
geneID = splicing_event
events = splicing_events[geneID]
index = 0
import collections
initial_group_psi_values = {}
try:
group_psi_values = collections.OrderedDict()
except Exception:
try:
import ordereddict
group_psi_values = ordereddict.OrderedDict()
except Exception:
group_psi_values = {}
for i in t[sampleIndexBegin:]: ### Value PSI range in the input file
try:
group = sample_group_index[index]
except Exception:
group = None
try:
try:
initial_group_psi_values[group].append([float(i), index])
except Exception:
initial_group_psi_values[group] = [[float(i), index]]
except Exception:
# print traceback.format_exc();sys.exit()
pass ### Ignore the NULL values
index += 1
### limit the number of events reported and sort based on the PSI values in each group
if "None" in groups and len(groups) == 1:
initial_group_psi_values["None"].sort()
group_size = len(initial_group_psi_values["None"]) / 2
filtered_group_index1 = map(lambda x: x[1], initial_group_psi_values["None"][:group_size])
filtered_group_index2 = map(lambda x: x[1], initial_group_psi_values["None"][group_size:])
group_psi_values["low"] = filtered_group_index1
group_psi_values["high"] = filtered_group_index2
else:
gn = 0
for group in groups:
gn += 1
# if gn>4: break
if group in initial_group_psi_values:
initial_group_psi_values[group].sort()
if len(groups) > 7:
filtered_group_indexes = map(lambda x: x[1], initial_group_psi_values[group][:1])
elif len(groups) > 5:
filtered_group_indexes = map(lambda x: x[1], initial_group_psi_values[group][:2])
elif len(groups) > 3:
filtered_group_indexes = map(lambda x: x[1], initial_group_psi_values[group][:4])
else:
filtered_group_indexes = map(lambda x: x[1], initial_group_psi_values[group][:5])
group_psi_values[group] = filtered_group_indexes
try:
update_plot_settings(bamdir, group_psi_values, sample_headers)
except Exception:
print "Cannot update the settings file. Likely permissions issue."
try:
reordered = reorderEvents([t[2] + " " + t[3]])
reordered = string.split(reordered[0], " ")
except Exception:
reordered = [t[2] + " " + t[3]]
reordered = string.split(reordered[0], " ")
# print reordered
if "PSI" in filename:
try:
formatted_splice_event = string.replace(reordered[1], ":", "__")
except Exception:
pass
### Submit the query
try:
ssp.plot_event(formatted_splice_event, index_dir, setting, outputdir)
success = True
except Exception:
success = False
# print traceback.format_exc()
else:
for event in events:
try:
ssp.plot_event(event, index_dir, setting, outputdir)
# print 'success' #formatted_splice_event='ENSMUSG00000000355__E4.1-E5.1'
except Exception: ### If it fails, output the gene-level plot
try:
ssp.plot_event(geneID, index_dir, setting, outputdir)
success = True
except Exception:
success = False
# print traceback.format_exc()
"""
### Second attempt
if 'PSI' in filename and success==False: ### Only relevant when parsing the junction pairs but not genes
try: formatted_splice_event=string.replace(reordered[0],':','__')
except Exception: pass
try: ssp.plot_event(formatted_splice_event,index_dir,setting,outputdir); # print 'success'
except Exception: pass
"""
return processed_events
def sashmi_plot_list(bamdir,fname,gene_label,lines,samp,gene_sym):
splicing_events=[]
type = None
firstLine = True
for line in open(fname,'rU').xreadlines():
line = cleanUpLine(line)
t = string.split(line,'\t')
if firstLine:
if 'junctionID-1' in t:
j1i = t.index('junctionID-1')
j2i = t.index('junctionID-2')
type='ASPIRE'
if 'ANOVA' in t:
type='PSI'
elif 'independent confirmation' in t:
type='confirmed'
elif 'ANOVA' in fname:
type = 'ANOVA'
firstLine=False
if ' ' in t[0] and ':' in t[0]:
splicing_events.append(t[0])
elif type=='ASPIRE':
splicing_events.append(t[j1i] +' '+ t[j2i])
elif type=='ANOVA':
try:
a,b = string.split(t[0],'|')
a = string.split(a,':')
a = string.join(a[1:],':')
splicing_events.append(a +' '+ b)
except Exception: pass
elif type=='PSI':
try:
j1,j2 = string.split(t[0],'|')
a,b,c = string.split(j1,':')
j1 = b+':'+c
splicing_events.append(j1 +' '+ j2)
except Exception:
#print traceback.format_exc();sys.exit()
pass
elif type=='confirmed':
try:
event_pair1 = string.split(t[1],'|')[0]
a,b,c,d = string.split(event_pair1,'-')
splicing_events.append(a+'-'+b +' '+ c+'-'+d)
except Exception: pass
if len(splicing_events)==0:
forceNoCompatibleEventsInFile
print 'Exporting plots',
for li in splicing_events:
if ":U" in li or "-U" in li:
continue
else:
li=cleanUpLine(li)
#print li
#dem[0]=['ENSG00000132424:I10.1 ENSG00000132424:E10.1-E11.1','ENSG00000146147:E10.3-E11.1 ENSG00000146147:E9.3-E15.1']
de=string.split(li,'\t')
dem[0]=de
#print dem[0]
for key in dem:
for i in range(len(dem[key])):
list1=[]
list2=[]
try:
k=gene_label.index(dem[key][i])
flag=1
lt=cleanUpLine(lines[k])
t=string.split(lt,'\t')
#print t
t=t[11:]
#print t
#list3=[]
#ind=[]
for x in range(len(t)):
#print x,t[x]
if(t[x]!=''):
if float(t[x]) < 0.8:
list1.append(x)
#print x
#print 'list1:'+str(x)
else:
list2.append(x)
#print x
# print str(x)
else:
continue
if len(list1)>5:
list1=list1[1:5]
if len(list2)>5:
list2=list2[1:5]
#print len(list1),len(list2)
except Exception:
for ij in range(len(samp)):
list1.append(ij)
update_plot_settings(bamdir,list1,list2,samp)
a=string.split(dem[key][i]," ")
if '-' in a[1]:
ch1=a[1]
f=string.split(a[0],':')
else:
ch1=a[0]
f=string.split(a[1],':')
event=findParentDir(inputpsi)
event=event+"trial_index/"
setting =unique.filepath("Config/sashimi_plot_settings.txt")
try: ch1=string.replace(ch1,':','__')
except Exception: pass
name=ch1
#outputdir=findParentDir(inputpsi)+"sashimiplots"
try: os.makedirs(outputdir)
except Exception: pass
#print '********',[ch1],[event],outputdir
try:
ssp.plot_event(ch1,event,setting,outputdir)
except Exception:
#print '^^^^^^^^^^^^',[ch1],[event],outputdir;sys.exit()
#print traceback.format_exc()
#print "error2"
#sys.exit()
continue
#outputdir=findParentDir(inputpsi)+"sashimiplots"
for filename in os.listdir(outputdir):
newname=string.split(filename,'/')
#print newname[0]
if newname[0] in gene_sym:
new_path = gene_sym[newname[0]]+'-'+filename
#new_path = string.replace()
os.rename(filename,new_path)
else:
continue
def sashmi_plot_list(bamdir,fname,gene_label,lines,samp,gene_sym):
line=fname.readlines()
fname.close()
for li in line:
if ":U" in li or "-U" in li:
continue
else:
li=cleanUpLine(li)
print li
#dem[0]=['ENSG00000132424:I10.1 ENSG00000132424:E10.1-E11.1','ENSG00000146147:E10.3-E11.1 ENSG00000146147:E9.3-E15.1']
de=string.split(li,'\t')
dem[0]=de
print dem[0]
for key in dem:
for i in range(len(dem[key])):
list1=[]
list2=[]
try:
k=gene_label.index(dem[key][i])
flag=1
lt=cleanUpLine(lines[k])
t=string.split(lt,'\t')
print t
t=t[11:]
print t
#list3=[]
#ind=[]
for x in range(len(t)):
print x,t[x]
if(t[x]!=''):
if float(t[x]) < 0.8:
list1.append(x)
#print x
#print 'list1:'+str(x)
else:
list2.append(x)
#print x
# print str(x)
else:
continue
if len(list1)>5:
list1=list1[1:5]
if len(list2)>5:
list2=list2[1:5]
print len(list1),len(list2)
except Exception:
for ij in range(len(samp)):
list1.append(ij)
update_plot_settings(bamdir,list1,list2,samp)
a=string.split(dem[key][i]," ")
if '-' in a[1]:
ch1=a[1]
f=string.split(a[0],':')
else:
ch1=a[0]
f=string.split(a[1],':')
event=findParentDir(inputpsi)
event=event+"trial_index/"
setting ="sashimi_plot_settings_2.txt"
name=ch1
outputdir=findParentDir(inputpsi)+"sashimiplots_aspire"
if not os.path.isdir(outputdir):
os.makedirs(outputdir)
try:
ssp.plot_event(ch1,event,setting,outputdir)
except Exception:
print "error2"
continue
outputdir=findParentDir(inputpsi)+"sashimiplots_aspire"
for filename in os.listdir(outputdir):
newname=string.split(filename,'/')
print newname[0]
if newname[0] in gene_sym:
os.rename(filename,gene_sym[newname[0]]+'-'+filename)
else:
continue
def sashmi_plot_list(bamdir, fname, gene_label, lines, samp, gene_sym):
splicing_events = []
type = None
firstLine = True
for line in open(fname, 'rU').xreadlines():
line = cleanUpLine(line)
t = string.split(line, '\t')
if firstLine:
if 'junctionID-1' in t:
j1i = t.index('junctionID-1')
j2i = t.index('junctionID-2')
type = 'ASPIRE'
if 'ANOVA' in t:
type = 'PSI'
elif 'independent confirmation' in t:
type == 'confirmed'
firstLine = False
if ' ' in t[0] and ':' in t[0]:
splicing_events.append(t[0])
elif type == 'ASPIRE':
splicing_events.append(t[j1i] + ' ' + t[j2i])
elif type == 'PSI':
try:
j1, j2 = string.split(t[0], '|')
a, b, c = string.split(j1, ':')
j1 = b + ':' + c
splicing_events.append(j1 + ' ' + j2)
except Exception:
pass
elif type == 'confirmed':
try:
event_pair1 = string.split(t[1], '|')[0]
a, b, c, d = string.split(event_pair1, '-')
splicing_events.append(a + '-' + b + ' ' + c + '-' + d)
except Exception:
pass
if len(splicing_events) == 0:
forceNoCompatibleEventsInFile
print 'Exporting plots',
for li in splicing_events:
if ":U" in li or "-U" in li:
continue
else:
li = cleanUpLine(li)
#print li
#dem[0]=['ENSG00000132424:I10.1 ENSG00000132424:E10.1-E11.1','ENSG00000146147:E10.3-E11.1 ENSG00000146147:E9.3-E15.1']
de = string.split(li, '\t')
dem[0] = de
#print dem[0]
for key in dem:
for i in range(len(dem[key])):
list1 = []
list2 = []
try:
k = gene_label.index(dem[key][i])
flag = 1
lt = cleanUpLine(lines[k])
t = string.split(lt, '\t')
#print t
t = t[11:]
#print t
#list3=[]
#ind=[]
for x in range(len(t)):
#print x,t[x]
if (t[x] != ''):
if float(t[x]) < 0.8:
list1.append(x)
#print x
#print 'list1:'+str(x)
else:
list2.append(x)
#print x
# print str(x)
else:
continue
if len(list1) > 5:
list1 = list1[1:5]
if len(list2) > 5:
list2 = list2[1:5]
#print len(list1),len(list2)
except Exception:
for ij in range(len(samp)):
list1.append(ij)
update_plot_settings(bamdir, list1, list2, samp)
a = string.split(dem[key][i], " ")
if '-' in a[1]:
ch1 = a[1]
f = string.split(a[0], ':')
else:
ch1 = a[0]
f = string.split(a[1], ':')
event = findParentDir(inputpsi)
event = event + "trial_index/"
setting = unique.filepath(
"Config/sashimi_plot_settings.txt")
name = ch1
#outputdir=findParentDir(inputpsi)+"sashimiplots"
try:
os.makedirs(outputdir)
except Exception:
pass
try:
ssp.plot_event(ch1, event, setting, outputdir)
except Exception:
#print traceback.format_exc()
#print "error2"
continue
#outputdir=findParentDir(inputpsi)+"sashimiplots"
for filename in os.listdir(outputdir):
newname = string.split(filename, '/')
#print newname[0]
if newname[0] in gene_sym:
new_path = gene_sym[newname[0]] + '-' + filename
#new_path = string.replace()
os.rename(filename, new_path)
else:
continue