def getAnnotations(Species, array_type, reannotate_exon_seq, force):
"""Annotate Affymetrix exon array data using files Ensembl data (sync'ed to genome release)."""
global species
species = Species
global test
global test_cluster
test = 'no'
test_cluster = ['TC0701360']
data_type = 'mRNA'
global ensembl_exon_db
global ensembl_exon_db
global exon_clusters
global exon_region_db
ensembl_exon_db, ensembl_annot_db, exon_clusters, intron_clusters, exon_region_db, intron_retention_db, ucsc_splicing_annot_db, ens_transcript_db = EnsemblImport.getEnsemblAssociations(
species, data_type, test)
ensembl_probeset_db = importCriticalExonLocations(
species, array_type, ensembl_exon_db,
force) ###Get Pre-computed genomic locations for critical exons
ensembl_probeset_db = ExonArrayEnsemblRules.annotateExons(
ensembl_probeset_db, exon_clusters, ensembl_exon_db, exon_region_db,
intron_retention_db, intron_clusters, ucsc_splicing_annot_db)
constitutive_gene_db = {}
ExonArrayEnsemblRules.exportEnsemblLinkedProbesets(array_type,
ensembl_probeset_db,
species)
print "\nCritical exon data exported coordinates, exon associations and splicing annotations exported..."
### Change filenames to reflect junction array type
export_filename = 'AltDatabase/' + species + '/' + array_type + '/' + species + '_Ensembl_probesets.txt'
ef = filepath(export_filename)
export_replacement = string.replace(export_filename, '_probe',
'_' + array_type + '_probe')
er = filepath(export_replacement)
shutil.copyfile(ef, er)
os.remove(ef) ### Copy file to a new name
### Export full exon seqeunce for probesets/critical exons to replace the original incomplete sequence (used for miRNA analyses)
if reannotate_exon_seq == 'yes':
JunctionArray.reAnnotateCriticalExonSequences(species, array_type)
def buildExonArrayExonAnnotations(species, array_type, force):
### Get UCSC associations (download databases if necessary)
mRNA_Type = 'mrna'; run_from_scratch = 'yes'
export_all_associations = 'no' ### YES only for protein prediction analysis
buildUCSCAnnoationFiles(species,mRNA_Type,export_all_associations,run_from_scratch,force)
from build_scripts import ExonArrayEnsemblRules; reload(ExonArrayEnsemblRules)
process_from_scratch='yes'
constitutive_source='default'
### Build the databases and return the variables (not used here)
source_biotype = 'mRNA'
if array_type == 'gene': source_biotype = 'gene'
probeset_db,annotate_db,constitutive_gene_db,splicing_analysis_db = ExonArrayEnsemblRules.getAnnotations(process_from_scratch,constitutive_source,source_biotype,species)
def displayExpressionGraph(species,
Platform,
exp_file,
gene,
transpose,
display=True,
showIntrons=False,
analysisType='plot'):
### Get gene annotations (users can provide an Ensembl or symbol)
print 'Importing exon-level expression data for visualization (be patient)...'
from build_scripts import ExonAnalyze_module
global platform
platform = Platform
if platform != 'AltMouse':
gene_annotation_file = "AltDatabase/ensembl/" + species + "/" + species + "_Ensembl-annotations.txt"
else:
gene_annotation_file = "AltDatabase/" + species + "/" + platform + "/" + platform + "_gene_annotations.txt"
genes = []
gene = string.replace(gene, '|', ',')
gene = string.replace(gene, ' ', ',')
if ',' in gene:
genes += string.split(gene, ',')
else:
genes.append(gene)
gene_db = {}
for gene in genes:
try:
if 'ENS' in gene:
try:
annotate_db ### If variable is defined
except Exception:
annotate_db = ExonAnalyze_module.import_annotations(
gene_annotation_file, platform,
keyBySymbol=False) ### Make an SQLite call
gene_symbol = annotate_db[gene].Symbol()
else:
try:
annotate_db ### If variable is defined
except Exception:
annotate_db = ExonAnalyze_module.import_annotations(
gene_annotation_file, platform, keyBySymbol=True)
gene_symbol = gene
gene = annotate_db[gene].GeneID()
gene_db[gene] = gene_symbol
except Exception:
#if len(gene)>0: print gene, 'not in database'
pass
if len(gene_db) == 0:
force_no_gene_found_error
if 'AltResults' in exp_file:
root_dir = string.split(exp_file, 'AltResults')[0] + 'ExonPlots/'
else:
root_dir = string.split(exp_file, 'ExpressionInput')[0] + 'ExonPlots/'
from build_scripts import ExonAnalyze_module
if platform == 'RNASeq': datatype = 'exons'
else: datatype = 'probesets'
export_exon_filename = 'AltDatabase/' + species + '/' + platform + '/' + species + '_Ensembl_' + datatype + '.txt'
if verifyFileLength(export_exon_filename) == 0:
rootdir = string.replace(root_dir, 'ExonPlots/', '')
export_exon_filename = rootdir + '/' + export_exon_filename
from build_scripts import ExonArrayEnsemblRules
ensembl_exon_db = ExonArrayEnsemblRules.reimportEnsemblProbesetsForSeqExtraction(
export_exon_filename, 'gene-probesets',
gene_db) ### Make an SQLite call
filter_db = {}
for gene in ensembl_exon_db:
ensembl_exon_db[gene].sort()
for (index, ed, id) in ensembl_exon_db[gene]:
filter_db[id] = []
try:
os.mkdir(root_dir)
except Exception:
None ### dir exists
print 'Image results being saved to the folder "ExonPlots" in the AltAnalyze results directory.'
importTableEntries(exp_file,
filter_db,
ensembl_exon_db,
gene_db,
root_dir,
transpose,
display,
showIntrons,
analysisType=analysisType) ### Make an SQLite call
def getJunctionComparisonsFromExport(species, array_type):
type = 'standard'
gene_junction_db = importEnsemblUCSCAltJunctions(species, type)
### Retrieve probesets with exon-junctions associated - these are critical exons
filename = 'AltDatabase/' + species + '/' + array_type + '/' + species + '_Ensembl_' + array_type + '_probesets.txt'
gene_probeset_db = ExonArrayEnsemblRules.reimportEnsemblProbesetsForSeqExtraction(
filename, 'junctions', {})
left = {}
right = {}
gene_db = {}
gene_exon_db = {}
nonjunction_aligning = {}
for gene in gene_probeset_db:
for (probe_data, ed) in gene_probeset_db[gene]:
probeset, strand, probeset_start, probeset_stop = probe_data
region_id = string.replace(ed.RegionNumber(), '-', '.')
original_region_id = region_id
region_ids = string.split(region_id, '|')
gene_db[probeset[:-2]] = gene
#ed.AssociatedSplicingJunctions()
r_starts = string.split(ed.ExonStart(), '|')
r_stops = string.split(ed.ExonStop(), '|')
for region_id in region_ids:
if '|5' in probeset:
try:
left[probeset[:-2]].append(region_id)
except Exception:
left[probeset[:-2]] = [region_id]
if strand == '+': ### If the junction probesets DO NOT align to the region coordinates, then the probeset maps to a junction outside the database
if probeset_stop not in r_stops:
nonjunction_aligning[
probeset[:
-2]] = original_region_id + '_' + probeset_stop, 'left'
elif probeset_start not in r_starts:
nonjunction_aligning[
probeset[:
-2]] = original_region_id + '_' + probeset_start, 'left'
elif '|3' in probeset:
try:
right[probeset[:-2]].append(region_id)
except Exception:
right[probeset[:-2]] = [region_id]
if strand == '+':
if probeset_start not in r_starts:
nonjunction_aligning[
probeset[:
-2]] = original_region_id + '_' + probeset_start, 'right'
elif probeset_stop not in r_stops:
nonjunction_aligning[
probeset[:
-2]] = original_region_id + '_' + probeset_stop, 'right'
else:
if '_' in region_id: print killer
try:
gene_exon_db[gene, region_id].append(probeset)
except Exception:
gene_exon_db[gene, region_id] = [probeset]
print 'len(nonjunction_aligning)', len(nonjunction_aligning)
gene_exon_db = eliminateRedundant(gene_exon_db)
junction_db = {} ### Get the exon-region IDs for an exon-junction
for probeset in left:
gene = gene_db[probeset]
if probeset in right:
for region1 in left[probeset]:
for region2 in right[probeset]:
junction = region1 + '-' + region2
try:
junction_db[gene, junction].append(probeset)
except Exception:
junction_db[gene, junction] = [probeset]
probeset_junction_export = 'AltDatabase/' + species + '/' + array_type + '/' + species + '_junction_comps.txt'
fn = filepath(probeset_junction_export)
data = open(fn, 'w')
print "Exporting", probeset_junction_export
title = 'gene' + '\t' + 'critical_exon' + '\t' + 'exclusion_junction_region' + '\t' + 'inclusion_junction_region' + '\t' + 'exclusion_probeset' + '\t' + 'inclusion_probeset' + '\t' + 'data_source' + '\n'
data.write(title)
temp_list = []
for (gene, critical_exon, incl_junction,
excl_junction) in gene_junction_db:
if (gene, incl_junction) in junction_db:
incl_junction_probesets = junction_db[gene, incl_junction]
if (gene, excl_junction) in junction_db:
excl_junction_probesets = junction_db[gene, excl_junction]
for incl_junction_probeset in incl_junction_probesets:
for excl_junction_probeset in excl_junction_probesets:
try:
for incl_exon_probeset in gene_exon_db[
gene, critical_exon]:
if incl_junction_probeset in nonjunction_aligning or excl_junction_probeset in nonjunction_aligning:
null = []
else: ### Ensure the probeset DOES map to the annotated junctions
temp_list.append(
string.join([
gene, critical_exon, excl_junction,
critical_exon,
excl_junction_probeset,
incl_exon_probeset, 'AltAnalyze'
], '\t') + '\n')
except Exception:
null = []
if incl_junction_probeset in nonjunction_aligning:
new_region_id, side = nonjunction_aligning[
incl_junction_probeset]
incl_junction = renameJunction(
incl_junction, side, new_region_id)
if excl_junction_probeset in nonjunction_aligning:
new_region_id, side = nonjunction_aligning[
excl_junction_probeset]
excl_junction = renameJunction(
excl_junction, side, new_region_id)
if excl_junction_probeset != incl_junction_probeset:
temp_list.append(
string.join([
gene, critical_exon, excl_junction,
incl_junction, excl_junction_probeset,
incl_junction_probeset, 'AltAnalyze'
], '\t') + '\n')
temp_list = unique.unique(temp_list)
for i in temp_list:
data.write(i)
data.close()
print 'Number of compared junctions exported', len(temp_list)