def load_objects(self, is_bedgraph, verbose=False):
"""Load files and initialize object"""
for i, t in enumerate(self.types):
if verbose:
print("Loading file ",
self.files[self.names[i]],
file=sys.stderr)
if t not in ["regions", "genes"] and verbose:
print("Cannot load objects", file=sys.stderr)
if t == "regions":
regions = GenomicRegionSet(self.names[i])
if is_bedgraph:
regions.read_bedgraph(
os.path.abspath(self.files[self.names[i]]))
else:
regions.read_bed(
os.path.abspath(self.files[self.names[i]])
) # Here change the relative path into absolute path
self.objectsDict[self.names[i]] = regions
elif t == "genes":
genes = GeneSet(self.names[i])
genes.read(
os.path.abspath(self.files[self.names[i]])
) # Here change the relative path into absolute path
self.objectsDict[self.names[i]] = genes
def load_objects(self, is_bedgraph, verbose=False, test=False):
"""Load files and initialize object.
*Keyword arguments:*
- is_bedgraph -- Whether regions are in bedgraph format (default = False).
- verbose -- Verbose output (default = False).
- test -- Fetch only 10 regions form each BED files for test.
"""
for i, t in enumerate(self.types):
if verbose:
print("Loading file ",
self.files[self.names[i]],
file=sys.stderr)
if t not in ["regions", "genes"] and verbose:
print("Cannot load objects", file=sys.stderr)
if t == "regions":
regions = GenomicRegionSet(self.names[i])
if is_bedgraph:
regions.read_bedgraph(
os.path.abspath(self.files[self.names[i]]))
else:
regions.read_bed(os.path.abspath(
self.files[self.names[i]]))
if test: regions.sequences = regions.sequences[0:11]
self.objectsDict[self.names[i]] = regions
elif t == "genes":
genes = GeneSet(self.names[i])
genes.read(
os.path.abspath(self.files[self.names[i]])
) # Here change the relative path into absolute path
self.objectsDict[self.names[i]] = genes
def load_objects(self, is_bedgraph, verbose=False, test=False):
"""Load files and initialize object.
*Keyword arguments:*
- is_bedgraph -- Whether regions are in bedgraph format (default = False).
- verbose -- Verbose output (default = False).
- test -- Fetch only 10 regions form each BED files for test.
"""
for i, t in enumerate(self.types):
if verbose: print("Loading file ", self.files[self.names[i]], file = sys.stderr)
if t not in ["regions", "genes"] and verbose:
print("Cannot load objects", file=sys.stderr)
if t == "regions":
regions = GenomicRegionSet(self.names[i])
if is_bedgraph:
regions.read_bedgraph(os.path.abspath(self.files[self.names[i]]))
else:
if test:
g = GenomicRegionSet(self.names[i])
g.read_bed(os.path.abspath(self.files[self.names[i]]))
regions.sequences = g.sequences[0:11]
else:
regions.read_bed(os.path.abspath(self.files[self.names[i]])) # Here change the relative path into absolute path
self.objectsDict[self.names[i]] = regions
elif t == "genes":
genes = GeneSet(self.names[i])
genes.read(os.path.abspath(self.files[self.names[i]])) # Here change the relative path into absolute path
self.objectsDict[self.names[i]] = genes
def match_ms_tags(self, field, test=False):
"""Add more entries to match the missing tags of the given field. For example, there are tags for cell like 'cell_A' and 'cell_B' for reads, but no these tag for regions. Then the regions are repeated for each tags from reads to match all reads.
*Keyword arguments:*
- field -- Field to add extra entries.
"""
# check regions or reads have empty tag
altypes = self.fieldsDict[field].keys()
if "ALL" in altypes:
altypes.remove("ALL")
for name in self.fieldsDict[field]["ALL"]:
i = self.names.index(name)
for t in altypes:
# print("\t"+t)
n = name + "_" + t
# print("\t\t"+n)
self.names.append(n)
self.types.append(self.types[i])
self.files[n] = self.files[name]
# types = self.get_types(name,skip_all=True)
# print("************")
# print(types)
for f in self.fields[3:]:
if f == field:
try:
self.fieldsDict[f][t].append(n)
except:
self.fieldsDict[f][t] = [n]
else:
try:
self.fieldsDict[f][self.get_type(
name=name, field=f)].append(n)
except:
self.fieldsDict[f][self.get_type(
name=name, field=f)] = [n]
# for f in self.fieldsDict.keys():
# for ty in types:
# try: self.fieldsDict[f][ty].append(n)
# except: pass
if self.types[i] == "regions":
g = GenomicRegionSet(n)
g.read_bed(self.files[name])
if test: g.sequences = g.sequences[0:11]
self.objectsDict[n] = g
self.trash.append(name)
def match_ms_tags(self,field):
"""Add more entries to match the missing tags of the given field. For example, there are tags for cell like 'cell_A' and 'cell_B' for reads, but no these tag for regions. Then the regions are repeated for each tags from reads to match all reads.
*Keyword arguments:*
- field -- Field to add extra entries.
"""
# print(field)
# print(self.fieldsDict)
# check regions or reads have empty tag
altypes = self.fieldsDict[field].keys()
if "ALL" in altypes:
altypes.remove("ALL")
for name in self.fieldsDict[field]["ALL"]:
# print(name)
i = self.names.index(name)
for t in altypes:
# print("\t"+t)
n = name+"_"+t
# print("\t\t"+n)
self.names.append(n)
self.types.append(self.types[i])
self.files[n] = self.files[name]
# types = self.get_types(name,skip_all=True)
# print("************")
# print(types)
for f in self.fields[3:]:
if f == field:
try: self.fieldsDict[f][t].append(n)
except: self.fieldsDict[f][t] = [n]
else:
try: self.fieldsDict[f][self.get_type(name=name,field=f)].append(n)
except: self.fieldsDict[f][self.get_type(name=name,field=f)] = [n]
# for f in self.fieldsDict.keys():
# for ty in types:
# try: self.fieldsDict[f][ty].append(n)
# except: pass
if self.types[i] == "regions":
g = GenomicRegionSet(n)
g.read_bed(self.files[name])
self.objectsDict[n] = g
self.trash.append(name)
##################################################################################
parser = argparse.ArgumentParser(description='Replace TCONs in BED file by assoicated gene names',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('-bed', type=str, help="BED file or a directory containing BED files")
parser.add_argument('-output', type=str, help="Define the output directory")
parser.add_argument('-organism', type=str, help="Define the organism")
args = parser.parse_args()
genome = GenomeData(args.organism)
if os.path.isfile(args.bed):
regionset = GenomicRegionSet("bed")
regionset.read_bed(args.bed)
gr = regionset.gene_association(organism=args.organism, promoterLength=1000,
threshDist=500000, show_dis=True)
regionset.replace_region_name(gr,combine=True)
regionset.write_bed(args.output)
elif os.path.isdir(args.bed):
if not os.path.exists(args.output):
os.makedirs(args.output)
for root, dirnames, filenames in os.walk(args.bed):
for filename in filenames:
if ".bed" in filename:
print(filename)
fnn = os.path.basename(filename)
from rgt.ExperimentalMatrix import *
#from fisher import pvalue
import scipy.stats
outdir = ""
back = False
designFile = sys.argv[1]
genomeName = sys.argv[2]
geneFile = sys.argv[3]
randomize = int(sys.argv[4])
backGroundPeaks = False
if len(sys.argv) > 5:
backGroundPeaksName = sys.argv[6]
backBed = GenomicRegionSet("BACK")
backBed.read_bed(backGroundPeaksName)
backGroundPeaks = True
distance = 50000
if len(sys.argv) > 6:
distance = len(sys.argv[6])
if len(sys.argv) > 7:
outdir = sys.argv[7]
#genomeFile=anotationPath+"chrom.sizes"
#geneFile=anotationPath+"association_file.bed"
exps = ExperimentalMatrix()
exps.read(designFile)
back=False
designFile = sys.argv[1]
anotationPath = sys.argv[2]
genomeFile=anotationPath+"chrom.sizes"
geneFile=anotationPath+"association_file.bed"
exps=ExperimentalMatrix()
exps.read(designFile)
beds=[]
geneLists=[]
#this should be improved
bedGenes = GenomicRegionSet(geneFile)
bedGenes.read_bed(geneFile)
allgenes=[]
for r in bedGenes:
allgenes.append(r.name)
allgenes=list(set(allgenes))
genesets=exps.get_genesets()
if len(sys.argv) > 3:
back=True
backGroundPeaks = sys.argv[3]
backBed=GenomicRegionSet("BACK")
backBed.read_bed(backGroundPeaks)
backBed=GenomicRegionSet("BACK")
if __name__ == "__main__":
import sys
from rgt.GenomicRegionSet import *
bam_file=sys.argv[1]
fasta_file=sys.argv[2]
bed_file=sys.argv[3]
kmer=int(sys.argv[4])
shift=int(sys.argv[5])
out=sys.argv[6]
regions=GenomicRegionSet("regions")
regions.read_bed(bed_file)
table=BiasTable(regions=regions,dnase_file_name=bam_file,genome_file_name=fasta_file,k_nb=kmer,shift=shift)
table.write_tables(out)
def load_exon_sequence(bed, directory, genome_path):
"""Load the exon sequence from the the transcripts.
Input BED format should contain:
blockCount - The number of blocks (exons) in the BED line.
blockSizes - A comma-separated list of the block sizes.
blockStarts - A comma-separated list of block starts.
see details: http://genome.ucsc.edu/FAQ/FAQformat#format1
Output:
Each FASTA file represants a transcript and contains all the exons within the file.
"""
regionset = GenomicRegionSet("bed")
regionset.read_bed(bed)
regionset.sort()
genome = pysam.Fastafile(genome_path)
try:
if len(regionset.sequences[0].data.split("\t")) == 7:
blockinfor = True
no_exon = False
except:
blockinfor = False
regionset.sequences.sort(key=lambda g: g.name)
no_exon = True
if blockinfor:
for gr in regionset:
if not gr.name:
print(
"Error: For fetching exon sequences, please define the transcript name."
)
sys.exit()
else:
if not os.path.exists(directory):
os.makedirs(directory)
f = open(os.path.join(directory, gr.name + ".fa"), "w")
data = gr.data.split("\t")
#print(len(data))
if len(data) == 7:
#print(data)
n = int(data[4])
blocks = [int(b) for b in filter(None, data[5].split(","))]
starts = [int(s) for s in filter(None, data[6].split(","))]
printstr = []
for i in range(n):
start = gr.initial + starts[i]
end = start + blocks[i]
if no_exon and i == 0:
ex = ""
elif gr.orientation == "-":
ex = "exon:" + str(n - i)
else:
ex = "exon:" + str(i + 1)
if gr.orientation == "-":
seq = Seq(
genome.fetch(gr.chrom, start - 1, end - 1),
IUPAC.unambiguous_dna)
seq = seq.reverse_complement()
p = [
">" + " ".join([
gr.name, ex, "_".join([
"REGION", gr.chrom,
str(start),
str(end), gr.orientation
])
]), seq
]
printstr.append(p)
else:
p = [
">" + " ".join([
gr.name, ex, "_".join([
"REGION", gr.chrom,
str(start),
str(end), gr.orientation
])
]),
genome.fetch(gr.chrom, start - 1, end - 1)
]
printstr.append(p)
if gr.orientation == "-": printstr = printstr[::-1]
for i in range(n):
print(printstr[i][0], file=f)
print(printstr[i][1], file=f)
else:
print(
"Warning: The given regions have no block information, please try write_bed_blocks"
)
f.close()
else:
pre_id = ""
for gr in regionset:
if not gr.name:
gr.name = gr.toString()
if pre_id == "":
pre_id = gr.name
z = GenomicRegionSet(gr.name)
z.add(gr)
elif gr.name == pre_id:
z.add(gr)
else:
f = open(os.path.join(directory, pre_id + ".fa"), "w")
for i, g in enumerate(z):
try:
regiontag = "_".join([
"REGION", g.chrom,
str(g.initial),
str(g.final), gr.orientation
])
except:
regiontag = "_".join(
["REGION", g.chrom,
str(g.initial),
str(g.final)])
print(">" + " ".join([g.name, regiontag]), file=f)
print(genome.fetch(g.chrom, g.initial, g.final), file=f)
f.close()
pre_id = gr.name
z = GenomicRegionSet(gr.name)
z.add(gr)
# Last TX
f = open(os.path.join(directory, pre_id + ".fa"), "w")
for i, g in enumerate(z):
try:
regiontag = "_".join([
"REGION", g.chrom,
str(g.initial),
str(g.final), gr.orientation
])
except:
regiontag = "_".join(
["REGION", g.chrom,
str(g.initial),
str(g.final)])
print(">" + " ".join([g.name, regiontag]), file=f)
print(genome.fetch(g.chrom, g.initial, g.final), file=f)
f.close()
def load_exon_sequence(bed, directory, genome_path):
"""Load the exon sequence from the the transcripts.
Input BED format should contain:
blockCount - The number of blocks (exons) in the BED line.
blockSizes - A comma-separated list of the block sizes.
blockStarts - A comma-separated list of block starts.
see details: http://genome.ucsc.edu/FAQ/FAQformat#format1
Output:
Each FASTA file represants a transcript and contains all the exons within the file.
"""
regionset = GenomicRegionSet("bed")
regionset.read_bed(bed)
regionset.sort()
genome = pysam.Fastafile(genome_path)
try:
if len(regionset.sequences[0].data.split("\t")) == 7:
blockinfor = True
no_exon = False
except:
blockinfor = False
regionset.sequences.sort(key=lambda g: g.name)
no_exon = True
if blockinfor:
for gr in regionset:
if not gr.name:
print("Error: For fetching exon sequences, please define the transcript name.")
sys.exit()
else:
if not os.path.exists(directory):
os.makedirs(directory)
f = open(os.path.join(directory, gr.name+".fa"), "w")
data = gr.data.split("\t")
#print(len(data))
if len(data) == 7:
#print(data)
n = int(data[4])
blocks = [ int(b) for b in filter(None, data[5].split(",")) ]
starts = [ int(s) for s in filter(None, data[6].split(",")) ]
printstr = []
for i in range(n):
start = gr.initial + starts[i]
end = start + blocks[i]
if no_exon and i == 0:
ex = ""
elif gr.orientation == "-":
ex = "exon:"+str(n-i)
else:
ex = "exon:"+str(i+1)
if gr.orientation == "-":
seq = Seq(genome.fetch(gr.chrom, start-1, end-1), IUPAC.unambiguous_dna)
seq = seq.reverse_complement()
p = [ ">"+ " ".join([ gr.name,
ex,
"_".join(["REGION",gr.chrom,
str(start),str(end),
gr.orientation]) ]),
seq ]
printstr.append(p)
else:
p = [ ">"+ " ".join([gr.name, ex,
"_".join(["REGION",gr.chrom,str(start),str(end), gr.orientation]) ]),
genome.fetch(gr.chrom, start-1, end-1)
]
printstr.append(p)
if gr.orientation == "-": printstr = printstr[::-1]
for i in range(n):
print(printstr[i][0], file=f)
print(printstr[i][1], file=f)
else:
print("Warning: The given regions have no block information, please try write_bed_blocks")
f.close()
else:
pre_id = ""
for gr in regionset:
if not gr.name:
gr.name = gr.toString()
if pre_id == "":
pre_id = gr.name
z = GenomicRegionSet(gr.name)
z.add(gr)
elif gr.name == pre_id:
z.add(gr)
else:
f = open(os.path.join(directory, pre_id+".fa"), "w")
for i, g in enumerate(z):
try: regiontag = "_".join(["REGION", g.chrom, str(g.initial),str(g.final), gr.orientation])
except: regiontag = "_".join(["REGION", g.chrom, str(g.initial),str(g.final)] )
print( ">"+ " ".join([g.name,
regiontag ]), file=f)
print(genome.fetch(g.chrom, g.initial, g.final), file=f)
f.close()
pre_id = gr.name
z = GenomicRegionSet(gr.name)
z.add(gr)
# Last TX
f = open(os.path.join(directory, pre_id+".fa"), "w")
for i, g in enumerate(z):
try: regiontag = "_".join(["REGION", g.chrom, str(g.initial),str(g.final), gr.orientation])
except: regiontag = "_".join(["REGION", g.chrom, str(g.initial),str(g.final)] )
print( ">"+ " ".join([g.name,
regiontag ]), file=f)
print(genome.fetch(g.chrom, g.initial, g.final), file=f)
f.close()