def justConvertFilenames(species, outputdir):
import gene_associations
gene_to_symbol = gene_associations.getGeneToUid(species,
('hide', 'Ensembl-Symbol'))
from import_scripts import OBO_import
symbol_to_gene = OBO_import.swapKeyValues(gene_to_symbol)
for filename in os.listdir(outputdir):
if '.pdf' in filename or '.png' in filename:
fn = string.replace(filename, '.pdf', '')
fn = string.replace(fn, '.png', '')
newname = string.split(fn, '__')
if newname[0] in gene_to_symbol:
new_filename = str(filename)
if '__' in filename:
new_filename = string.split(filename, '__')[1]
elif '\\' in filename:
new_filename = string.split(filename, '\\')[1]
elif '/' in filename:
new_filename = string.split(filename, '/')[1]
nnname = gene_to_symbol[
newname[0]][0] + '-SashimiPlot_' + new_filename
try:
os.rename(os.path.join(outputdir, filename),
os.path.join(outputdir, nnname))
except Exception:
pass
else:
continue
def importGeneSymbols(species):
import gene_associations
gene_to_symbol = gene_associations.getGeneToUid(species,
('hide', 'Ensembl-Symbol'))
from import_scripts import OBO_import
symbol_to_gene = OBO_import.swapKeyValues(gene_to_symbol)
return gene_to_symbol, symbol_to_gene
def sashmi_plot_list(bamdir,
eventsToVisualizeFilename,
PSIFilename,
events=None):
try:
import gene_associations
gene_to_symbol = gene_associations.getGeneToUid(
species, ('hide', 'Ensembl-Symbol'))
from import_scripts import OBO_import
symbol_to_gene = OBO_import.swapKeyValues(gene_to_symbol)
except Exception:
symbol_to_gene = {}
if events == None:
splicing_events, expandedSearch = importSplicingEventsToVisualize(
eventsToVisualizeFilename)
else:
### Replace any ":" from the input events
#for i in range(len(events)): events[i] = string.replace(events[i],':','__')
expandedSearch = True
for i in range(len(events)):
gene = string.split(events[i], '__')[0]
if gene in gene_to_symbol:
symbol = gene_to_symbol[gene][0]
elif 'ENS' not in gene or 'G0000' in gene:
if gene in symbol_to_gene:
ensID = symbol_to_gene[gene][0]
symbol = gene
events[
i] = ensID ### translate this ID to an Ensembl gene ID for propper SashimiPlot lookup
splicing_events = events ### optionally get from supplied variable
if len(splicing_events) == 0:
print eventsToVisualizeFilename
forceNoCompatibleEventsInFile
print 'Exporting plots',
### Determine Groups for Coloring
groups_file = 'None'
dir_list = unique.read_directory(root_dir + '/ExpressionInput')
for file in dir_list:
if 'groups.' in file:
groups_file = root_dir + '/ExpressionInput/' + file
if groups_file != None:
try:
import ExpressionBuilder
sample_group_db = ExpressionBuilder.simplerGroupImport(groups_file)
groups = []
for sample in sample_group_db:
if sample_group_db[sample] not in groups:
groups.append(sample_group_db[sample]
) ### create an ordered list of unique group
except Exception:
groups = ['None']
#print traceback.format_exc()
pass
processed_events = formatAndSubmitSplicingEventsToSashimiPlot(
PSIFilename, bamdir, splicing_events, sample_group_db, groups, False)
mopup_events = getMopUpEvents(splicing_events, processed_events)
### Do the same for supplied gene queries or junctions that didn't map above using the gene expression values as a guide
#print len(splicing_events),len(processed_events),len(mopup_events)
processed_events = formatAndSubmitSplicingEventsToSashimiPlot(
steady_state_exp_file, bamdir, mopup_events, sample_group_db, groups,
expandedSearch)
if len(processed_events) > 0:
mopup_events = getMopUpEvents(mopup_events, processed_events)
processed_events = formatAndSubmitSplicingEventsToSashimiPlot(
PSIFilename, bamdir, mopup_events, sample_group_db, groups, True)
return gene_to_symbol
def importDataSimple(filename, input_type, MOD=None, Species=None):
id_db = {}
fn = filepath(filename)
x = 0
for line in open(fn, 'rU').xreadlines():
data = cleanUpLine(line)
t = string.split(data, '\t')
if data[0] == '#' and x == 0: x = 0
elif x == 0:
column_headers = t[1:]
if input_type != 'LineageProfiler':
try:
column_headers = t[
2] ### exclude the ID, system code and p-value column headers
except Exception:
column_headers = 'NA'
x = 1
else:
if x == 1 and input_type != 'LineageProfiler':
### get system conversions
system_code = t[1]
import GO_Elite
from import_scripts import OBO_import
system_codes, source_types, mod_types = GO_Elite.getSourceData(
)
source_data = system_codes[system_code]
try:
Mod = mod ### global established in upstream functions
speciescode = species_code
except Exception:
Mod = MOD
speciescode = Species
if source_data == Mod:
source_is_mod = True
else:
source_is_mod = False
mod_source = Mod + '-' + source_data + '.txt'
gene_to_source_id = gene_associations.getGeneToUid(
speciescode, ('hide', mod_source))
source_to_gene = OBO_import.swapKeyValues(
gene_to_source_id)
if input_type != 'LineageProfiler':
if source_is_mod == True:
try:
id_db[t[0]] = float(t[2])
except Exception:
id_db[t[0]] = 'NA'
elif t[0] in source_to_gene:
mod_ids = source_to_gene[t[0]]
for mod_id in mod_ids:
try:
value = t[2]
except Exception:
value = 'NA'
if value == '+': value = 1
elif value == '-': value = -1
try:
id_db[mod_id] = float(
value
) ### If multiple Ensembl IDs in dataset, only record the last associated fold change
except Exception:
id_db[mod_id] = 'NA'
break
else:
id_db[t[0]] = map(float, t[1:]) ### Applies to LineageProfiler
x += 1
#print len(id_db),column_headers
return id_db, column_headers
def reformatPolyAdenylationCoordinates(species, force):
""" PolyA annotations are currently only available from UCSC for human, but flat file
annotations from 2003-2006 are available for multiple species. Convert these to BED format"""
version = {}
version['Rn'] = '2003(rn3)'
version['Dr'] = '2003(zv4)'
version['Gg'] = '2004(galGal2)'
version['Hs'] = '2006(hg8)'
version['Mm'] = '2004(mm5)'
print 'Exporting polyADB_2 coordinates as BED for', species
### Obtain the necessary database files
url = 'http://altanalyze.org/archiveDBs/all/polyAsite.txt'
output_dir = 'AltDatabase/ucsc/' + species + '/'
if force == 'yes':
filename, status = update.download(url, output_dir, '')
else:
filename = output_dir + 'polyAsite.txt'
### Import the refseq to Ensembl information
import gene_associations
from import_scripts import OBO_import
from build_scripts import EnsemblImport
import export
try:
ens_unigene = gene_associations.getGeneToUid(species,
'Ensembl-UniGene')
print len(ens_unigene), 'Ensembl-UniGene entries imported'
external_ensembl = OBO_import.swapKeyValues(ens_unigene)
use_entrez = 'no'
except Exception:
ens_entrez = gene_associations.getGeneToUid(species,
'Ensembl-EntrezGene')
print len(ens_entrez), 'Ensembl-EntrezGene entries imported'
external_ensembl = OBO_import.swapKeyValues(ens_entrez)
use_entrez = 'yes'
gene_location_db = EnsemblImport.getEnsemblGeneLocations(
species, 'RNASeq', 'key_by_array')
export_bedfile = output_dir + species + '_polyADB_2_predictions.bed'
print 'exporting', export_bedfile
export_data = export.ExportFile(export_bedfile)
header = '#' + species + '\t' + 'polyADB_2' + '\t' + version[species] + '\n'
export_data.write(header)
fn = filepath(filename)
x = 0
not_found = {}
for line in open(fn, 'rU').xreadlines():
data = cleanUpLine(line)
if x == 0: x = 1
else:
siteid, llid, chr, sitenum, position, supporting_EST, cleavage = string.split(
data, '\t')
if chr == 'chrM':
chr = 'chrMT' ### MT is the Ensembl convention whereas M is the Affymetrix and UCSC convention
if chr == 'M':
chr = 'MT' ### MT is the Ensembl convention whereas M is the Affymetrix and UCSC convention
if species in siteid:
if 'NA' not in chr: chr = 'chr' + chr
strand = '+'
geneid = siteid
pos_start = str(int(position) - 1)
pos_end = position
if use_entrez == 'no':
external_geneid = string.join(
string.split(siteid, '.')[:2], '.')
else:
external_geneid = llid
if external_geneid in external_ensembl:
ens_geneid = external_ensembl[external_geneid][0]
geneid += '-' + ens_geneid
chr, strand, start, end = gene_location_db[ens_geneid]
else:
not_found[external_geneid] = []
bed_format = string.join(
[chr, pos_start, pos_end, geneid, '0', '-'], '\t'
) + '\n' ### We don't know the strand, so write out both strands
export_data.write(bed_format)
bed_format = string.join(
[chr, pos_start, pos_end, geneid, '0', strand],
'\t') + '\n'
export_data.write(bed_format)
export_data.close()