def buildJunctionExonAnnotations(species, array_type, specific_array_type,
force, genomic_build):
### Get UCSC associations (download databases if necessary)
mRNA_Type = 'mrna'
run_from_scratch = 'yes'
force = 'no'
export_all_associations = 'no' ### YES only for protein prediction analysis
buildUCSCAnnoationFiles(species, mRNA_Type, export_all_associations,
run_from_scratch, force)
### Get genomic locations and initial annotations for exon sequences (exon pobesets and junctions)
import JunctionArray
import JunctionArrayEnsemblRules
""" The following functions:
1) Extract transcript cluster-to-gene annotations
2) Extract exon sequences for junctions and exon probesets from the Affymetrix annotation file (version 2.0),
3) Map these sequences to Ensembl gene sequences (build specific) plus and minus 2KB, upstream and downstream
4) Obtain AltAnalyze exon region annotations and obtain full-length exon sequences for each exon probeset
5) Consoladate these into an Ensembl_probeset.txt file (rather than Ensembl_junction_probeset.txt) with junctions
having a single probeset identifier.
6) Determine which junctions and junction-exons represent recipricol junctions using:
a) AltAnalyze identified recipricol junctions from Ensembl and UCSC and
b) Affymetrix suggested recipricol junctions based on common exon cluster annotations, creating
Mm_junction_comps_updated.txt.
c) De novo comparison of all exon-junction region IDs for all junctions using the EnsemblImport method compareJunctions().
"""
### Steps 1-3
JunctionArray.getJunctionExonLocations(species, array_type,
specific_array_type)
### Step 4
JunctionArrayEnsemblRules.getAnnotations(species, array_type, 'yes', force)
### Step 5-6
JunctionArray.identifyJunctionComps(species, array_type,
specific_array_type)
def buildJunctionExonAnnotations(species,array_type,specific_array_type,force,genomic_build):
### Get UCSC associations (download databases if necessary)
mRNA_Type = 'mrna'; run_from_scratch = 'yes'; force='no'
export_all_associations = 'no' ### YES only for protein prediction analysis
#buildUCSCAnnoationFiles(species,mRNA_Type,export_all_associations,run_from_scratch,force)
### Get genomic locations and initial annotations for exon sequences (exon pobesets and junctions)
import JunctionArray
import JunctionArrayEnsemblRules
""" The following functions:
1) Extract transcript cluster-to-gene annotations
2) Extract exon sequences for junctions and exon probesets from the Affymetrix annotation file (version 2.0),
3) Map these sequences to Ensembl gene sequences (build specific) plus and minus 2KB, upstream and downstream
4) Obtain AltAnalyze exon region annotations and obtain full-length exon sequences for each exon probeset
5) Consoladate these into an Ensembl_probeset.txt file (rather than Ensembl_junction_probeset.txt) with junctions
having a single probeset identifier.
6) Determine which junctions and junction-exons represent recipricol junctions using:
a) AltAnalyze identified recipricol junctions from Ensembl and UCSC and
b) Affymetrix suggested recipricol junctions based on common exon cluster annotations, creating
Mm_junction_comps_updated.txt.
c) De novo comparison of all exon-junction region IDs for all junctions using the EnsemblImport method compareJunctions().
"""
### Steps 1-3
JunctionArray.getJunctionExonLocations(species,array_type,specific_array_type)
### Step 4
JunctionArrayEnsemblRules.getAnnotations(species,array_type,'yes',force)
### Step 5-6
JunctionArray.identifyJunctionComps(species,array_type,specific_array_type)
def buildAltMouseExonAnnotations(species,array_type,force,genomic_build):
"""Code required to:
1) Extract out Affymetrix provided exon sequence (probeset sequence extracted from "probeset_sequence_reversed.txt", derived
directly from the Affymetrix AltMouse probe annotation file), from the "SEQUENCE-transcript-dbase.txt" (built using
dump-chip1 .gff sequence and AltMerge-Peptide Informatics script "sequence_analysis_AltMouse_refseq.py").
2) Once exported, grab full length exon sequences using exon/intron coordinates matches to full-length gene sequences with 2kb
flanking sequence to efficiently predict microRNA binding site exclusion (reAnnotateCriticalExonSequences) and later for
coordinate mapping to get exons aligning with UCSC annotated splicing annotations and exons. This sequence data replaced
the previous file (don't need to re-run this - see rederive_exonseq == 'yes' below for reference).
3) Match the updated exon sequences to the most recent genomic coordinates and build the exact equivalent of the exon array
Mm_Ensembl_probeset.txt database (same structure and ExonArrayEnsemblRules.py code). This involves running EnsemblImport.
This code should be run before the exon array location build code since the "Mm_Ensembl_probeset.txt" is created and then re-
written as "Mm_AltMouse_Ensembl_probeset.txt".
"""
import JunctionArray
import JunctionArrayEnsemblRules
rederive_exonseq = 'no'
### Only needs to be run once, to export exon sequence for AltMouse array the original (1 and 2 above)
if rederive_exonseq == 'yes':
import AltAnalyze
import ExonAnnotate_module
import ExonAnalyze_module
agglomerate_inclusion_probesets = 'no'; onlyAnalyzeJunctions='no'
probeset_annotations_file = "AltDatabase/"+species+"/"+array_type+"/"+"MASTER-probeset-transcript.txt"
verifyFile(probeset_annotations_file,array_type) ### Will force download if missing
exon_db={}; filtered_arrayids={};filter_status='no'
constituitive_probeset_db,exon_db,genes_being_analyzed = AltAnalyze.importSplicingAnnotationDatabase(probeset_annotations_file,array_type,filtered_arrayids,filter_status)
alt_junction_db,critical_exon_db,exon_dbase,exon_inclusion_db,exon_db = ExonAnnotate_module.identifyPutativeSpliceEvents(exon_db,constituitive_probeset_db,{},agglomerate_inclusion_probesets,onlyAnalyzeJunctions)
ExonAnnotate_module.exportJunctionComparisons(alt_junction_db,critical_exon_db,exon_dbase)
print "Finished exporting junctions used in AltMouse array comparisons."
ExonAnalyze_module.exportAltMouseExonSequence()
JunctionArray.reAnnotateCriticalExonSequences(species,array_type)
### 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)
reannotate_exon_seq = 'yes'
print 'genomic_build', genomic_build
if genomic_build == 'new':
### Need to run with every new genomic build (match up new coordinates
print "Begining to derive exon sequence from new genomic build"
JunctionArray.identifyCriticalExonLocations(species,array_type)
reannotate_exon_seq = 'yes'
JunctionArrayEnsemblRules.getAnnotations(species,array_type,reannotate_exon_seq,force)
### Download files required during AltAnalyze analysis but not during the database build process
filename = "AltDatabase/"+species+"/"+array_type+"/"+"MASTER-probeset-transcript.txt"
verifyFile(filename,array_type) ### Will force download if missing
filename = "AltDatabase/"+species+'/'+ array_type+'/'+array_type+"_annotations.txt"
verifyFile(filename,array_type) ### Will force download if missing
def importCriticalExonLocations(species,array_type,ensembl_exon_db,force):
###ensembl_exon_db[(geneid,chr,strand)] = [[E5,exon_info]] #exon_info = (exon_start,exon_stop,exon_id,exon_annot)
###ensembl_probeset_db[geneid,chr,strand].append(probeset_data) #probeset_data = [start,stop,probeset_id,exon_class,transcript_cluster_id]
gene_info_db = {}
for (ens_geneid,chr,strand) in ensembl_exon_db: gene_info_db[ens_geneid] = chr,strand
filename = 'AltDatabase/'+species+'/'+array_type+'/'+array_type+'_critical_exon_locations.txt'
array_ensembl={}
###Get the most recent gene-symbol annotations (applicable with a new Ensembl build for the same genomic build)
ensembl_symbol_db = getEnsemblAnnotations(species)
primary_gene_annotation_file = 'AltDatabase/'+species +'/'+ array_type +'/'+ array_type+ '_gene_annotations.txt'
update.verifyFile(primary_gene_annotation_file,array_type)
array_gene_annotations = JunctionArray.importGeneric(primary_gene_annotation_file)
for array_geneid in array_gene_annotations:
t = array_gene_annotations[array_geneid]; description=t[0];entrez=t[1];symbol=t[2]
if symbol in ensembl_symbol_db and len(symbol)>0 and len(array_geneid)>0:
ens_geneid = ensembl_symbol_db[symbol]
if len(ens_geneid)>0: array_ensembl[array_geneid]= ens_geneid
update.verifyFile(filename,array_type)
ensembl_probeset_db = importJunctionLocationData(filename,array_ensembl,gene_info_db,test)
print len(ensembl_probeset_db), "Genes inlcuded in",array_type,"location database"
return ensembl_probeset_db
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 executeParameters(species,array_type,force,genomic_build,update_uniprot,update_ensembl,update_probeset_to_ensembl,update_domain,update_miRs,update_all,update_miR_seq,ensembl_version):
if '|' in array_type: array_type, specific_array_type = string.split(array_type,'|') ### To destinguish between array sub-types, like the HJAY and hGlue
else: specific_array_type = array_type
if update_all == 'yes':
update_uniprot='yes'; update_ensembl='yes'; update_probeset_to_ensembl='yes'; update_domain='yes'; update_miRs = 'yes'
if update_ensembl == 'yes':
import EnsemblSQL; reload(EnsemblSQL)
""" Used to grab all essential Ensembl annotations previously obtained via BioMart"""
configType = 'Advanced'; analysisType = 'AltAnalyzeDBs'; externalDBName = ''
EnsemblSQL.buildEnsemblRelationalTablesFromSQL(species,configType,analysisType,externalDBName,ensembl_version,force)
""" Used to grab Ensembl-to-External gene associations"""
configType = 'Basic'; analysisType = 'ExternalOnly'; externalDBName = 'Uniprot/SWISSPROT'
EnsemblSQL.buildEnsemblRelationalTablesFromSQL(species,configType,analysisType,externalDBName,ensembl_version,force)
""" Used to grab Ensembl full gene sequence plus promoter and 3'UTRs """
if array_type == 'AltMouse' or array_type == 'junction' or array_type == 'RNASeq':
EnsemblSQL.getFullGeneSequences(ensembl_version,species)
if update_uniprot == 'yes':
###Might need to delete the existing versions of downloaded databases or force download
buildUniProtFunctAnnotations(species,force)
if update_probeset_to_ensembl == 'yes':
if species == 'Mm' and array_type == 'AltMouse':
buildAltMouseExonAnnotations(species,array_type,force,genomic_build)
elif array_type == 'junction':
buildJunctionExonAnnotations(species,array_type,specific_array_type,force,genomic_build)
elif array_type == 'RNASeq':
import RNASeq; test_status = 'no'; data_type = 'mRNA'
RNASeq.getEnsemblAssociations(species,data_type,test_status,force)
else: buildExonArrayExonAnnotations(species,array_type,force)
if update_domain == 'yes':
### Get UCSC associations for all Ensembl linked genes (download databases if necessary) if species == 'Mm' and array_type == 'AltMouse':
mRNA_Type = 'mrna'; run_from_scratch = 'yes'
export_all_associations = 'yes' ### YES only for protein prediction analysis
buildUCSCAnnoationFiles(species,mRNA_Type,export_all_associations,run_from_scratch,force)
if (species == 'Mm' and array_type == 'AltMouse'):
"""Imports and re-exports array-Ensembl annotations"""
import JunctionArray
null = JunctionArray.importArrayAnnotations(species,array_type); null={}
if (species == 'Mm' and array_type == 'AltMouse') or array_type == 'junction' or array_type == 'RNASeq':
"""Performs probeset sequence aligment to Ensembl and UCSC transcripts. To do: Need to setup download if files missing"""
import mRNASeqAlign; analysis_type = 'reciprocal'
mRNASeqAlign.alignProbesetsToTranscripts(species,array_type,analysis_type,force)
import IdentifyAltIsoforms; run_seqcomp = 'no'
IdentifyAltIsoforms.runProgram(species,array_type,'null',force,run_seqcomp)
import FeatureAlignment; import JunctionArray
FeatureAlignment.findDomainsByGenomeCoordinates(species,array_type,'null')
if array_type == 'junction' or array_type == 'RNASeq':
### For junction probeset sequences from mRNASeqAlign(), find and assess alternative proteins - export to the folder 'junction'
mRNASeqAlign.alignProbesetsToTranscripts(species,array_type,'single',force)
IdentifyAltIsoforms.runProgram(species,array_type,'junction',force,run_seqcomp)
FeatureAlignment.findDomainsByGenomeCoordinates(species,array_type,'junction')
### For exon probesets (and junction exons) align and assess alternative proteins - export to the folder 'exon'
IdentifyAltIsoforms.runProgram(species,array_type,'exon',force,run_seqcomp)
# FeatureAlignment.findDomainsByGenomeCoordinates(species,array_type,'exon') # not needed
""" Repeat above with CoordinateBasedMatching = True """
### Peform coordinate based junction mapping to transcripts (requires certain sequence files built in IdentifyAltIosofmrs)
analysis_type = 'reciprocal'
mRNASeqAlign.alignProbesetsToTranscripts(species,array_type,analysis_type,force,CoordinateBasedMatching = True)
IdentifyAltIsoforms.runProgram(species,array_type,'null',force,run_seqcomp)
FeatureAlignment.findDomainsByGenomeCoordinates(species,array_type,'null')
mRNASeqAlign.alignProbesetsToTranscripts(species,array_type,'single',force,CoordinateBasedMatching = True)
IdentifyAltIsoforms.runProgram(species,array_type,'junction',force,run_seqcomp)
FeatureAlignment.findDomainsByGenomeCoordinates(species,array_type,'junction')
IdentifyAltIsoforms.runProgram(species,array_type,'exon',force,run_seqcomp)
if array_type == 'RNASeq':
JunctionArray.combineExonJunctionAnnotations(species,array_type)
if update_miRs == 'yes':
if update_miR_seq == 'yes':
import MatchMiRTargetPredictions; only_add_sequence_to_previous_results = 'no'
MatchMiRTargetPredictions.runProgram(species,force,only_add_sequence_to_previous_results)
if array_type == 'exon' or array_type == 'gene':
import ExonSeqModule
stringency = 'strict'; process_microRNA_predictions = 'yes'; mir_source = 'multiple'
ExonSeqModule.runProgram(species,array_type,process_microRNA_predictions,mir_source,stringency)
stringency = 'lax'
ExonSeqModule.runProgram(species,array_type,process_microRNA_predictions,mir_source,stringency)
ExonArray.exportMetaProbesets(array_type,species) ### Export metaprobesets for this build
else:
import JunctionSeqModule
stringency = 'strict'; mir_source = 'multiple'
JunctionSeqModule.runProgram(species,array_type,mir_source,stringency,force)
stringency = 'lax'
JunctionSeqModule.runProgram(species,array_type,mir_source,stringency,force)
if array_type == 'junction':
try:
import JunctionArray; import JunctionArrayEnsemblRules
JunctionArray.filterForCriticalExons(species,array_type)
JunctionArray.overRideExonEntriesWithJunctions(species,array_type)
JunctionArrayEnsemblRules.annotateJunctionIDsAsExon(species,array_type)
ExonArray.exportMetaProbesets(array_type,species) ### Export metaprobesets for this build
except IOError: print 'No built junction files to analyze';sys.exit()
if array_type == 'RNASeq' and (species == 'Hs' or species == 'Mm' or species == 'Rn'):
import JunctionArray; import JunctionArrayEnsemblRules
try: JunctionArrayEnsemblRules.annotateJunctionIDsAsExon(species,array_type)
except IOError: print 'No Ensembl_exons.txt file to analyze';sys.exit()
try:
filename = 'AltDatabase/'+species+'/SequenceData/miRBS-combined_gene-targets.txt'; ef=filepath(filename)
er = string.replace(ef,species+'/SequenceData/miRBS-combined_gene-targets.txt','ensembl/'+species+'/'+species+'_microRNA-Ensembl.txt')
import shutil; shutil.copyfile(ef,er)
except Exception: null=[]
if array_type != 'RNASeq':
### Get the probeset-probe relationships from online - needed for FIRMA analysis
filename = 'AltDatabase/'+species+'/'+array_type+'/'+species+'_probeset-probes.txt'
if array_type == 'junction' and 'lue' in specific_array_type:
server_folder = 'junction/hGlue'
verifyFile(filename,server_folder) ### Will force download if missing
verifyFile('AltDatabase/'+species+'/'+array_type+'/platform.txt',server_folder) ### Will force download if missing
elif array_type != 'AltMouse': verifyFile(filename,array_type) ### Will force download if missing
if (array_type == 'exon' or array_type == 'AltMouse') and species != 'Rn':
try:
### Available for select exon-arrays and AltMouse
probeset_to_remove_file = 'AltDatabase/'+species+'/'+array_type+'/'+species+'_probes_to_remove.txt'
verifyFile(probeset_to_remove_file,array_type)
except Exception: null=[]
output_file = 'AltDatabase/'+species+'/'+array_type+'/'+species+'_'+array_type+'-exon_probesets.txt'
fn=filepath(output_file); data = open(fn,'w')
data.write(array_type+'_probeset\texon_probeset\n')
for probeset in junction_to_exonids:
exon_probeset = junction_to_exonids[probeset]
data.write(probeset+'\t'+exon_probeset+'\n')
data.close()
if __name__ == '__main__':
m = 'Mm'; h = 'Hs'
Species = m
array_type = 'RNASeq' ###In theory, could be another type of junciton or combination array
annotateJunctionIDsAsExon(Species,array_type); sys.exit()
#reimportJunctionComps(Species,array_type,'original');kill
#JunctionArray.getJunctionExonLocations(Species,array_type)
"""
### Get UCSC associations (download databases if necessary)
import UCSCImport
mRNA_Type = 'mrna'; run_from_scratch = 'yes'; force='no'
export_all_associations = 'no' ### YES only for protein prediction analysis
UCSCImport.runUCSCEnsemblAssociations(Species,mRNA_Type,export_all_associations,run_from_scratch,force)
"""
getAnnotations(Species,array_type,'yes','no')
JunctionArray.identifyJunctionComps(Species,array_type)
#importAndReformatEnsemblJunctionAnnotations(Species,array_type)
def executeParameters(species, array_type, force, genomic_build,
update_uniprot, update_ensembl,
update_probeset_to_ensembl, update_domain, update_miRs,
update_all, update_miR_seq, ensembl_version):
if '|' in array_type:
array_type, specific_array_type = string.split(
array_type, '|'
) ### To destinguish between array sub-types, like the HJAY and hGlue
else:
specific_array_type = array_type
if update_all == 'yes':
update_uniprot = 'yes'
update_ensembl = 'yes'
update_probeset_to_ensembl = 'yes'
update_domain = 'yes'
update_miRs = 'yes'
if update_ensembl == 'yes':
import EnsemblSQL
reload(EnsemblSQL)
""" Used to grab all essential Ensembl annotations previously obtained via BioMart"""
configType = 'Advanced'
analysisType = 'AltAnalyzeDBs'
externalDBName = ''
EnsemblSQL.buildEnsemblRelationalTablesFromSQL(species, configType,
analysisType,
externalDBName,
ensembl_version, force)
""" Used to grab Ensembl-to-External gene associations"""
configType = 'Basic'
analysisType = 'ExternalOnly'
externalDBName = 'Uniprot/SWISSPROT'
EnsemblSQL.buildEnsemblRelationalTablesFromSQL(species, configType,
analysisType,
externalDBName,
ensembl_version, force)
""" Used to grab Ensembl full gene sequence plus promoter and 3'UTRs """
if array_type == 'AltMouse' or array_type == 'junction' or array_type == 'RNASeq':
EnsemblSQL.getFullGeneSequences(ensembl_version, species)
if update_uniprot == 'yes':
###Might need to delete the existing versions of downloaded databases or force download
buildUniProtFunctAnnotations(species, force)
if update_probeset_to_ensembl == 'yes':
if species == 'Mm' and array_type == 'AltMouse':
buildAltMouseExonAnnotations(species, array_type, force,
genomic_build)
elif array_type == 'junction':
buildJunctionExonAnnotations(species, array_type,
specific_array_type, force,
genomic_build)
elif array_type == 'RNASeq':
import RNASeq
test_status = 'no'
data_type = 'mRNA'
RNASeq.getEnsemblAssociations(species, data_type, test_status,
force)
else:
buildExonArrayExonAnnotations(species, array_type, force)
if update_domain == 'yes':
### Get UCSC associations for all Ensembl linked genes (download databases if necessary) if species == 'Mm' and array_type == 'AltMouse':
mRNA_Type = 'mrna'
run_from_scratch = 'yes'
export_all_associations = 'yes' ### YES only for protein prediction analysis
buildUCSCAnnoationFiles(species, mRNA_Type, export_all_associations,
run_from_scratch, force)
if (species == 'Mm' and array_type == 'AltMouse'):
"""Imports and re-exports array-Ensembl annotations"""
import JunctionArray
null = JunctionArray.importArrayAnnotations(species, array_type)
null = {}
if (species == 'Mm' and array_type == 'AltMouse'
) or array_type == 'junction' or array_type == 'RNASeq':
"""Performs probeset sequence aligment to Ensembl and UCSC transcripts. To do: Need to setup download if files missing"""
import mRNASeqAlign
analysis_type = 'reciprocal'
mRNASeqAlign.alignProbesetsToTranscripts(species, array_type,
analysis_type, force)
import IdentifyAltIsoforms
run_seqcomp = 'no'
IdentifyAltIsoforms.runProgram(species, array_type, 'null', force,
run_seqcomp)
import FeatureAlignment
FeatureAlignment.findDomainsByGenomeCoordinates(
species, array_type, 'null')
if array_type == 'junction' or array_type == 'RNASeq':
### For junction probeset sequences from mRNASeqAlign(), find and assess alternative proteins - export to the folder 'junction'
mRNASeqAlign.alignProbesetsToTranscripts(species, array_type,
'single', force)
IdentifyAltIsoforms.runProgram(species, array_type, 'junction',
force, run_seqcomp)
FeatureAlignment.findDomainsByGenomeCoordinates(
species, array_type, 'junction')
if array_type == 'junction' or array_type == 'RNASeq':
### For exon probesets (and junction exons) align and assess alternative proteins - export to the folder 'exon'
IdentifyAltIsoforms.runProgram(species, array_type, 'exon',
force, run_seqcomp)
# FeatureAlignment.findDomainsByGenomeCoordinates(species,array_type,'exon') # not needed
if array_type == 'RNASeq':
import JunctionArray
JunctionArray.combineExonJunctionAnnotations(
species, array_type)
if update_miRs == 'yes':
if update_miR_seq == 'yes':
import MatchMiRTargetPredictions
only_add_sequence_to_previous_results = 'no'
MatchMiRTargetPredictions.runProgram(
species, force, only_add_sequence_to_previous_results)
if array_type == 'exon' or array_type == 'gene':
import ExonSeqModule
stringency = 'strict'
process_microRNA_predictions = 'yes'
mir_source = 'multiple'
ExonSeqModule.runProgram(species, array_type,
process_microRNA_predictions, mir_source,
stringency)
stringency = 'lax'
ExonSeqModule.runProgram(species, array_type,
process_microRNA_predictions, mir_source,
stringency)
ExonArray.exportMetaProbesets(
array_type, species) ### Export metaprobesets for this build
else:
import JunctionSeqModule
stringency = 'strict'
mir_source = 'multiple'
JunctionSeqModule.runProgram(species, array_type, mir_source,
stringency, force)
stringency = 'lax'
JunctionSeqModule.runProgram(species, array_type, mir_source,
stringency, force)
if array_type == 'junction':
try:
import JunctionArray
import JunctionArrayEnsemblRules
JunctionArray.filterForCriticalExons(species, array_type)
JunctionArray.overRideExonEntriesWithJunctions(species, array_type)
JunctionArrayEnsemblRules.annotateJunctionIDsAsExon(
species, array_type)
ExonArray.exportMetaProbesets(
array_type, species) ### Export metaprobesets for this build
except IOError:
print 'No built junction files to analyze'
sys.exit()
if array_type == 'RNASeq' and (species == 'Hs' or species == 'Mm'
or species == 'Rn'):
import JunctionArray
import JunctionArrayEnsemblRules
try:
JunctionArrayEnsemblRules.annotateJunctionIDsAsExon(
species, array_type)
except IOError:
print 'No Ensembl_exons.txt file to analyze'
sys.exit()
try:
filename = 'AltDatabase/' + species + '/SequenceData/miRBS-combined_gene-targets.txt'
ef = filepath(filename)
er = string.replace(
ef, species + '/SequenceData/miRBS-combined_gene-targets.txt',
'ensembl/' + species + '/' + species + '_microRNA-Ensembl.txt')
import shutil
shutil.copyfile(ef, er)
except Exception:
null = []
if array_type != 'RNASeq':
### Get the probeset-probe relationships from online - needed for FIRMA analysis
filename = 'AltDatabase/' + species + '/' + array_type + '/' + species + '_probeset-probes.txt'
if array_type == 'junction' and 'lue' in specific_array_type:
server_folder = 'junction/hGlue'
verifyFile(filename,
server_folder) ### Will force download if missing
verifyFile('AltDatabase/' + species + '/' + array_type +
'/platform.txt',
server_folder) ### Will force download if missing
elif array_type != 'AltMouse':
verifyFile(filename,
array_type) ### Will force download if missing
if (array_type == 'exon'
or array_type == 'AltMouse') and species != 'Rn':
try:
### Available for select exon-arrays and AltMouse
probeset_to_remove_file = 'AltDatabase/' + species + '/' + array_type + '/' + species + '_probes_to_remove.txt'
verifyFile(probeset_to_remove_file, array_type)
except Exception:
null = []
def buildAltMouseExonAnnotations(species, array_type, force, genomic_build):
"""Code required to:
1) Extract out Affymetrix provided exon sequence (probeset sequence extracted from "probeset_sequence_reversed.txt", derived
directly from the Affymetrix AltMouse probe annotation file), from the "SEQUENCE-transcript-dbase.txt" (built using
dump-chip1 .gff sequence and AltMerge-Peptide Informatics script "sequence_analysis_AltMouse_refseq.py").
2) Once exported, grab full length exon sequences using exon/intron coordinates matches to full-length gene sequences with 2kb
flanking sequence to efficiently predict microRNA binding site exclusion (reAnnotateCriticalExonSequences) and later for
coordinate mapping to get exons aligning with UCSC annotated splicing annotations and exons. This sequence data replaced
the previous file (don't need to re-run this - see rederive_exonseq == 'yes' below for reference).
3) Match the updated exon sequences to the most recent genomic coordinates and build the exact equivalent of the exon array
Mm_Ensembl_probeset.txt database (same structure and ExonArrayEnsemblRules.py code). This involves running EnsemblImport.
This code should be run before the exon array location build code since the "Mm_Ensembl_probeset.txt" is created and then re-
written as "Mm_AltMouse_Ensembl_probeset.txt".
"""
import JunctionArray
import JunctionArrayEnsemblRules
rederive_exonseq = 'no'
### Only needs to be run once, to export exon sequence for AltMouse array the original (1 and 2 above)
if rederive_exonseq == 'yes':
import AltAnalyze
import ExonAnnotate_module
import ExonAnalyze_module
agglomerate_inclusion_probesets = 'no'
onlyAnalyzeJunctions = 'no'
probeset_annotations_file = "AltDatabase/" + species + "/" + array_type + "/" + "MASTER-probeset-transcript.txt"
verifyFile(probeset_annotations_file,
array_type) ### Will force download if missing
exon_db = {}
filtered_arrayids = {}
filter_status = 'no'
constituitive_probeset_db, exon_db, genes_being_analyzed = AltAnalyze.importSplicingAnnotationDatabase(
probeset_annotations_file, array_type, filtered_arrayids,
filter_status)
alt_junction_db, critical_exon_db, exon_dbase, exon_inclusion_db, exon_db = ExonAnnotate_module.identifyPutativeSpliceEvents(
exon_db, constituitive_probeset_db, {},
agglomerate_inclusion_probesets, onlyAnalyzeJunctions)
ExonAnnotate_module.exportJunctionComparisons(alt_junction_db,
critical_exon_db,
exon_dbase)
print "Finished exporting junctions used in AltMouse array comparisons."
ExonAnalyze_module.exportAltMouseExonSequence()
JunctionArray.reAnnotateCriticalExonSequences(species, array_type)
### 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)
reannotate_exon_seq = 'yes'
print 'genomic_build', genomic_build
if genomic_build == 'new':
### Need to run with every new genomic build (match up new coordinates
print "Begining to derive exon sequence from new genomic build"
JunctionArray.identifyCriticalExonLocations(species, array_type)
reannotate_exon_seq = 'yes'
JunctionArrayEnsemblRules.getAnnotations(species, array_type,
reannotate_exon_seq, force)
### Download files required during AltAnalyze analysis but not during the database build process
filename = "AltDatabase/" + species + "/" + array_type + "/" + "MASTER-probeset-transcript.txt"
verifyFile(filename, array_type) ### Will force download if missing
filename = "AltDatabase/" + species + '/' + array_type + '/' + array_type + "_annotations.txt"
verifyFile(filename, array_type) ### Will force download if missing
