def build_transcript_data(species, gene_bed, gene_mrna, gene_pre_mrna, pre_mrna):
"""
Generates transcript data structures to call peaks on
Allows for either predefined files (from the data directory)
or custom files
Accepts species, and genebed, genemrnaand genepremrna options
species - the species to run on
gene_bed - an abribtary bed file of locations to search for peaks (should be gene locations)
gene_mrna - the effective length of the mrna of a gene (unmappable regions removed)
gene_premrna - the effective length of the pre-mrna (unmappable regions removed)
pre_mrna - flag True indicates use pre-mRNA lengths instead of mRNA lengths
returns genes and lengths dict
"""
#error checking
acceptable_species = get_acceptable_species()
if (species is None and
gene_bed is None and
(gene_mrna is None or gene_pre_mrna is None)):
raise ValueError("You must set either \"species\" or \"geneBed\"+\"geneMRNA\"+\"genePREMRNA\"")
if species is not None and gene_bed is not None:
raise ValueError("You shouldn't set both geneBed and species, defaults exist for %s" % (acceptable_species))
#Now actually assign values
if species is not None:
try:
gene_bed = clipper.data_file(species + ".AS.STRUCTURE_genes.BED.gz")
gene_mrna = clipper.data_file(species + ".AS.STRUCTURE_mRNA.lengths")
gene_pre_mrna = clipper.data_file(species + ".AS.STRUCTURE_premRNA.lengths")
except ValueError:
raise ValueError("Defaults don't exist for your species: %s. Please choose from: %s or supply \"geneBed\"+\"geneMRNA\"+\"genePREMRNA\"" % (species, acceptable_species))
#Selects mRNA or preMRNA lengths
if pre_mrna is True:
lenfile = gene_pre_mrna
else:
lenfile = gene_mrna
if lenfile is None:
raise IOError("""didn't pass correct mRNA length file option
with given length file""")
#builds dict to do processing on,
genes = build_geneinfo(gene_bed)
lengths = build_lengths(lenfile)
return genes, lengths
def build_transcript_data_gtf_as_structure(species, pre_mrna):
"""
gtf_file - gtf file generated from AS_STRUCTURE_gtf ipython notebook
pre_mrna - if true uses pre mRNA length instead of mRNA length
"""
bedtoolintervals = []
x = clipper.data_file(species + ".AS.STRUCTURE.COMPILED.gff")
gtf_file = pybedtools.BedTool(x)
for gene in gtf_file:
effective_length = gene.attrs[
'premrna_length'] if pre_mrna else gene.attrs['mrna_length']
attrs = "gene_id=%s;" % (gene.attrs['gene_id'])
if "transcript_ids" in gene.attrs:
attrs += "transcript_ids=%s;" % (gene.attrs['transcript_ids'])
attrs += "effective_length=%s" % (str(effective_length))
bedtoolintervals.append(
pybedtools.create_interval_from_list(
map(str, [
gene['chrom'], "AS_STRUCTURE", "mRNA",
str(gene.start + 1),
str(gene.stop + 1), "0", gene['strand'], ".", attrs
])))
return pybedtools.BedTool(bedtoolintervals)
def build_transcript_data_gtf_as_structure(species, pre_mrna):
"""
gtf_file - gtf file generated from AS_STRUCTURE_gtf ipython notebook
pre_mrna - if true uses pre mRNA length instead of mRNA length
"""
results = []
x = clipper.data_file(species + ".AS.STRUCTURE.COMPILED.gff")
gtf_file = pybedtools.BedTool(x)
for gene in gtf_file:
effective_length = gene.attrs['premrna_length'] if pre_mrna else gene.attrs['mrna_length']
attrs = "gene_id=%s;" % (gene.attrs['gene_id'])
if "transcript_ids" in gene.attrs:
attrs += "transcript_ids=%s;" % (gene.attrs['transcript_ids'])
attrs += "effective_length=%s" % (str(effective_length))
results.append(pybedtools.create_interval_from_list(map(str, [gene['chrom'],
"AS_STRUCTURE",
"mRNA",
str(gene.start + 1),
str(gene.stop + 1),
"0",
gene['strand'],
".",
attrs
])))
return pybedtools.BedTool(results)
def build_transcript_data_gtf_as_structure(species, pre_mrna):
"""
calculate effective length for each transcript from pre-created gtf file in clipper/data
Returns Bedtool containing effective length
:param species: (str) genome name
:param pre_mrna: (bool) if true uses pre-mRNA length instead of mRNA length
:return: (pybedtools.Bedtool)
:rtype: pybedtools.BedTool
"""
bedtool_intervals = []
x = clipper.data_file(species + ".AS.STRUCTURE.COMPILED.gff")
gtf_file = pybedtools.BedTool(x)
for gene in gtf_file:
effective_length = gene.attrs['premrna_length'] if pre_mrna else gene.attrs['mrna_length']
attrs = "gene_id=%s;" % (gene.attrs['gene_id'])
if "transcript_ids" in gene.attrs:
attrs += "transcript_ids=%s;" % (gene.attrs['transcript_ids'])
attrs += "effective_length=%s" % (str(effective_length))
# add to bedtool_intervals
to_string = map(str, [gene['chrom'],"AS_STRUCTURE","mRNA",str(gene.start + 1),str(gene.stop + 1),"0",gene['strand'],".",attrs]) # map object
bedtool_intervals.append(pybedtools.create_interval_from_list(list(to_string)))
return pybedtools.BedTool(bedtool_intervals)
def build_transcript_data(species, gene_bed, gene_mrna, gene_pre_mrna,
pre_mrna):
"""
Generates transcript data structures to call peaks on
Allows for either predefined files (from the data directory)
or custom files
Accepts species, and genebed, genemrnaand genepremrna options
species - the species to run on
gene_bed - an abribtary bed file of locations to search for peaks (should be gene locations)
gene_mrna - the effective length of the mrna of a gene (unmappable regions removed)
gene_premrna - the effective length of the pre-mrna (unmappable regions removed)
pre_mrna - flag True indicates use pre-mRNA lengths instead of mRNA lengths
returns genes and lengths dict
"""
#error checking
acceptable_species = get_acceptable_species()
if (species is None and gene_bed is None
and (gene_mrna is None or gene_pre_mrna is None)):
raise ValueError(
"You must set either \"species\" or \"geneBed\"+\"geneMRNA\"+\"genePREMRNA\""
)
if species is not None and gene_bed is not None:
raise ValueError(
"You shouldn't set both geneBed and species, defaults exist for %s"
% (acceptable_species))
#Now actually assign values
if species is not None:
try:
gene_bed = clipper.data_file(species +
".AS.STRUCTURE_genes.BED.gz")
gene_mrna = clipper.data_file(species +
".AS.STRUCTURE_mRNA.lengths")
gene_pre_mrna = clipper.data_file(species +
".AS.STRUCTURE_premRNA.lengths")
except ValueError:
raise ValueError(
"Defaults don't exist for your species: %s. Please choose from: %s or supply \"geneBed\"+\"geneMRNA\"+\"genePREMRNA\""
% (species, acceptable_species))
#Selects mRNA or preMRNA lengths
if pre_mrna is True:
lenfile = gene_pre_mrna
else:
lenfile = gene_mrna
if lenfile is None:
raise IOError("""didn't pass correct mRNA length file option
with given length file""")
#builds dict to do processing on,
genes = build_geneinfo(gene_bed)
lengths = build_lengths(lenfile)
#this is a stopgap until it can be fully factored out, returing a gtf file of
#genes and effective lengths, eventually this is the file we want to pass in
gtf_list = []
for gene in genes.keys():
gtf_list.append(
pybedtools.create_interval_from_list([
genes[gene][0], "AS_STRUCTURE", "mRNA",
str(genes[gene][2]),
str(genes[gene][3]), ".",
str(genes[gene][4]), ".",
"gene_id=" + gene + "; effective_length=" + str(lengths[gene])
]))
return pybedtools.BedTool(gtf_list)
def build_transcript_data(species, gene_bed, gene_mrna, gene_pre_mrna, pre_mrna):
"""
Generates transcript data structures to call peaks on
Allows for either predefined files (from the data directory)
or custom files
Accepts species, and genebed, genemrnaand genepremrna options
species - the species to run on
gene_bed - an abribtary bed file of locations to search for peaks (should be gene locations)
gene_mrna - the effective length of the mrna of a gene (unmappable regions removed)
gene_premrna - the effective length of the pre-mrna (unmappable regions removed)
pre_mrna - flag True indicates use pre-mRNA lengths instead of mRNA lengths
returns genes and lengths dict
"""
#error checking
acceptable_species = get_acceptable_species()
if (species is None and
gene_bed is None and
(gene_mrna is None or gene_pre_mrna is None)):
raise ValueError("You must set either \"species\" or \"geneBed\"+\"geneMRNA\"+\"genePREMRNA\"")
if species is not None and gene_bed is not None:
raise ValueError("You shouldn't set both geneBed and species, defaults exist for %s" % (acceptable_species))
#Now actually assign values
if species is not None:
try:
gene_bed = clipper.data_file(species + ".AS.STRUCTURE_genes.BED.gz")
gene_mrna = clipper.data_file(species + ".AS.STRUCTURE_mRNA.lengths")
gene_pre_mrna = clipper.data_file(species + ".AS.STRUCTURE_premRNA.lengths")
except ValueError:
raise ValueError("Defaults don't exist for your species: %s. Please choose from: %s or supply \"geneBed\"+\"geneMRNA\"+\"genePREMRNA\"" % (species, acceptable_species))
#Selects mRNA or preMRNA lengths
if pre_mrna is True:
lenfile = gene_pre_mrna
else:
lenfile = gene_mrna
if lenfile is None:
raise IOError("""didn't pass correct mRNA length file option
with given length file""")
#builds dict to do processing on,
genes = build_geneinfo(gene_bed)
lengths = build_lengths(lenfile)
#this is a stopgap until it can be fully factored out, returing a gtf file of
#genes and effective lengths, eventually this is the file we want to pass in
gtf_list = []
for gene in genes.keys():
gtf_list.append(pybedtools.create_interval_from_list([genes[gene][0],
"AS_STRUCTURE",
"mRNA",
str(genes[gene][2]),
str(genes[gene][3]),
".",
str(genes[gene][4]),
".",
"gene_id=" + gene + "; effective_length=" + str(lengths[gene])]))
return pybedtools.BedTool(gtf_list)
def build_transcript_data(species, gene_bed, gene_mrna, gene_pre_mrna,
pre_mrna):
"""
Generates transcript data structures to call peaks on
Allows for either predefined files (from the data directory)
or custom files
Accepts species, and genebed, genemrnaand genepremrna options
species - the species to run on
gene_bed - an abribtary bed file of locations to search for peaks (should be gene locations)
gene_mrna - the effective length of the mrna of a gene (unmappable regions removed)
gene_premrna - the effective length of the pre-mrna (unmappable regions removed)
returns genes and lengths dict
"""
#error checking
acceptable_species = get_acceptable_species()
if (species is None and gene_bed is None
and (gene_mrna is None or gene_pre_mrna is None)):
raise ValueError(
"You must set either \"species\" or \"geneBed\"+\"geneMRNA\"+\"genePREMRNA\""
)
if species is not None and gene_bed is not None:
raise ValueError(
"You shouldn't set both geneBed and species, defaults exist for %s"
% (acceptable_species))
#Now actually assign values
if species is not None:
try:
gene_bed = clipper.data_file(species +
".AS.STRUCTURE_genes.BED.gz")
gene_mrna = clipper.data_file(species +
".AS.STRUCTURE_mRNA.lengths")
gene_pre_mrna = clipper.data_file(species +
".AS.STRUCTURE_premRNA.lengths")
except ValueError:
raise ValueError(
"Defaults don't exist for your species: %s. Please choose from: %s or supply \"geneBed\"+\"geneMRNA\"+\"genePREMRNA\""
% (species, acceptable_species))
#Selects mRNA or preMRNA lengths
if pre_mrna is True:
lenfile = gene_pre_mrna
else:
lenfile = gene_mrna
if lenfile is None:
raise IOError("""didn't pass correct mRNA length file option
with given length file""")
#builds dict to do processing on,
genes = build_geneinfo(gene_bed)
lengths = build_lengths(lenfile)
return genes, lengths