def __init__(self, genomeBuild="hg18"):
self.sequenceBase = os.path.expanduser("~/mount/publicdata/" +
genomeBuild +
"/assembly/fasta/")
self.chromosomeEnds = ChromosomeEnds(genomeBuild)
self.chrms = {}
def __init__(self, build, annotation, exons, minExons, numbExons, tsses,
chunks, ud, ui, dd, di):
assert not (exons and tsses)
#Stores all parameters
self.exons = exons
self.minExons = minExons
self.numbExons = numbExons
self.tsses = tsses
self.geneNumbChunks = 20 if (chunks == None) else chunks
self.upstreamDistance = 5000 if (ud == None) else ud
self.upstreamInterval = 1000 if (ui == None) else ui
self.downstreamDistance = 5000 if (dd == None) else dd
self.downstreamInterval = 1000 if (di == None) else di
# Gets dictionaries of genes, exons transcripts etc from Ensembl class
#assert build == "hg18", "Non hg-18 not supported for Ensembl regions at the moment"
#Gets the lengths of chromosomes
# we keep this as a non instance variable as well so that the subclass below can access it
genedata = Ensembl.EnsemblGenes(assembly=build, annotation=annotation)
self.genedata = genedata
self.chromosomeEnds = ChromosomeEnds(build)
# Gets an object of the genelist class
class EnsemblNameAndIDFormatter(list):
def __init__(self, genesToUseLocation):
geneList = GeneList(genesToUseLocation)
self.seengenes = set()
for gene in geneList:
if gene in self.seengenes:
# seen this gene id before (generally shouldnt be the case as using GeneList which guarantees this
# for the source list at least
pass
elif gene in genedata:
# we've not seen the gene but it's in the ensembl ids list
self.append(gene)
self.seengenes.add(gene)
else:
# it's not in the ensembl ids list, it could be a gene name
found = False
for geneid in genedata.getGeneIDs(gene):
if geneid not in self.seengenes:
self.append(geneid)
self.seengenes.add(geneid)
found = True
if not found:
print "No GeneID for:" + gene
print genesToUseLocation + ":" + str(len(self))
self.regionIterator = EnsemblNameAndIDFormatter
def __init__(self, build, ud, ui, dd, di):
self.upstreamDistance = 100000 if (ud == None) else ud
self.upstreamInterval = 1000 if (ui == None) else ui
self.downstreamDistance = 100000 if (dd == None) else dd
self.downstreamInterval = 1000 if (di == None) else di
self.chromosomeEnds = ChromosomeEnds(build)
self.regionIterator = PointList
def __init__(self, build, chunks, ud, ui, dd, di):
self.numbChunks = 20 if (chunks == None) else chunks
self.upstreamDistance = 100000 if (ud == None) else ud
self.upstreamInterval = 1000 if (ui == None) else ui
self.downstreamDistance = 100000 if (dd == None) else dd
self.downstreamInterval = 1000 if (di == None) else di
print self.numbChunks, self.upstreamDistance, self.upstreamInterval, self.downstreamDistance, self.downstreamInterval
self.chromosomeEnds = ChromosomeEnds(build)
self.regionIterator = BedList
print str(err) # will print something like "option -a not recognized"
sys.exit(2)
# add executing directory as part of path
sys.path.append(sys.path[0])
direction = True
assembly = "hg18"
for opt, arg in opts:
if opt == "-a":
a = arg
elif opt == "-b":
b = arg
elif opt == "--assembly":
assembly = arg
gaps = ChromosomeGaps(assembly)
ends = ChromosomeEnds(assembly)
effectiveGenomeSize = 0
for chr in ChrList(assembly):
effectiveGenomeSize += ends[chr] - gaps.chrmgaps[chr]
print "---"
print a.split("/")[-1] + " which have " + b.split("/")[-1]
print "---"
compareTwo(a, b, effectiveGenomeSize, ChrList(assembly))
import gc
gc.disable()
from bed.treatment import SimpleBed
from bed.treatment import Bed as BedIntervalTree
from datastructures.genomeintervaltree import GenomeIntervalTree
from genemapping.gaps import ChromosomeGaps
from genemapping.chrmEnds import ChromosomeEnds
from genemapping.chrList import ChrList
from multiprocessing import Pool
import random
import collections
gaps = ChromosomeGaps("hg18")
ends = ChromosomeEnds("hg18")
from datastructures.wrg import WeightedRandomGenerator
#weights = collections.defaultdict(int)
effectiveGenomeSize = 0
for chr in ChrList("hg18"):
effectiveGenomeSize += ends[chr] - gaps.chrmgaps[chr]
# below is needed for weighted chromosome selection
#weights[chr] = ends[chr] - gaps.chrmgaps[chr]
#assert weights[chr] > 0
print "Effective Genome Size:" + str(effectiveGenomeSize)
# optional - removed for now. allows regions to move between chromosomes
#randomChromosome = WeightedRandomGenerator(weights)
sys.exit(2)
# add executing directory as part of path
sys.path.append(sys.path[0])
vstepWidth = 200 #default
pointlist = False
for o, a in opts:
if (o=="-s"):
vstepWidth = int(a)
elif (o=="-d"):
debug = True
elif (o=="-c"):
chromosomeEnds = ChromosomeEnds(a)
elif (o=="-g"):
GeneSummary.surroundingDistance = int(a)
for o, a in opts:
if o == "-s":
pass # we dealt with this already
elif (o=="-d"):
pass # we dealt with this already
elif (o=="-c"):
pass # we dealt with this already
elif (o=="-g"):
pass # we dealt with this already
# gene mappings / point list settings (for how to deal with the arguments representing which genes / points to plot)
elif o == "-a":
def __init__(self,filename,build):
self.treatment = BedFile(filename)
self.getValues = self.treatment.getIntervalsInRange
self.chromosomeEnds = ChromosomeEnds(build)
def __init__(self,build):
self.genome = Genome(genomeBuild = build)
self.valuesBehaviour = missingValuesDontCount
self.chromosomeEnds = ChromosomeEnds(build)
joinedfile = a
print "Joined file: ", a
elif o == "--binsize":
binsize = int(a)
print "Binsize: ", a
#assert False, "Two different chromosomes not implemented yet"
elif o == "--genome":
genome = a
print "Genome: ", genome
elif o == "--minsize":
minsize = int(a)
print "Minsize: ", minsize
assert joinedfile != None
chrmEnds = ChromosomeEnds(genome)
chrmList = ChrList(genome)
chrmMatrix = defaultdict(dict)
for chrmA in chrmList:
for chrmB in chrmList:
chrmMatrix[chrmA][chrmB] = defaultdict(int)
chrmMatrix[chrmB][chrmA] = defaultdict(int)
reads = csv.reader(open(joinedfile, "r"), delimiter="\t")
for read in reads:
#print read[lread_chr],read[lread_pos],read[rread_chr],read[rread_pos]
lread_chr = read[lread_chr_col]
rread_chr = read[rread_chr_col]
outfile = fname + ".extended"
print "writing to outfile", outfile
writer = csv.writer(open(outfile, "w"), delimiter="\t")
for row in reader:
chr = row[0]
start = int(row[1])
end = int(row[2])
strand = row[5]
if strand == "+":
row[2] = min(ends[chr], start + extendlen)
elif strand == "-":
row[1] = max(0, end - extendlen)
writer.writerow(row)
if __name__ == "__main__":
from glob import glob
import sys
if len(sys.argv) != 4:
print "usage: extendbed.py <beddir> <genomebuild> <extendlength>"
sys.exit(1)
ends = ChromosomeEnds(sys.argv[2])
beddir = sys.argv[1]
extendlen = int(sys.argv[3])
allfiles = glob(beddir + "/*.bed")
for fname in allfiles:
extendOne(fname, extendlen)