rearrangement = False
# deletion = Interval("chr" + chr_num, start, end)
write_all_chrms_in_file = False #set True if you want write training file consisting several chromosomes
fill_empty_contacts = False #set True if you want use all contacts in region, without empty contacts
logging.getLogger(__name__).debug("Using input folder " + input_folder)
# Read contacts data
genome = fastaReader(args['path_to_genome'],
useOnlyChromosomes=[chromosome]) #str(chr_num)])
genome = genome.read_data()
print(genome.data)
now = datetime.datetime.now()
params.contacts_reader = hicReader(fname=input_folder + "/" + cell_type +
"/" + hic_name,
genome=genome,
binsize=params.binsize)
params.contacts_reader = params.contacts_reader.read_data(
fill_empty_contacts=fill_empty_contacts, noDump=False)
if params.use_only_contacts_with_CTCF == "cont_with_CTCF":
params.proportion = 1
params.contacts_reader.use_contacts_with_CTCF(CTCFfile=input_folder+"/" + cell_type+"/CTCF/"+CTCF_file_name,
maxdist=params.maxdist,
proportion=params.proportion,
keep_only_orient=params.keep_only_orient,
CTCForientfile=input_folder + "/" + cell_type + \
"/CTCF/"+CTCF_file_name+"-orient.bed")
params.use_only_contacts_with_CTCF += str(
params.contacts_reader.conts_with_ctcf)
#make deletion
params.sample_size = 100 # how many contacts write to file
params.conttype = conttype
params.max_cpus = 11
params.keep_only_orient = False # set True if you want use only CTCF with orient
#params.use_only_contacts_with_CTCF = "cont_with_CTCF" # "cont_with_CTCF"
params.use_only_contacts_with_CTCF = "no"
# use this option to change proportion
# of contacts with nearest ctcf sites in training datasets
write_all_chrms_in_file = True # set True if you have train with few chromosomes. Need for writing different chromosomes in the same file
fill_empty_contacts = False
logging.getLogger(__name__).debug("Using input folder " + input_folder)
# Read contacts data
params.contacts_reader = ContactsReader()
contacts_files = []
# set path to the coefficient file and to contacts files
# contacts file format: bin_start--bin_end--contact_count
[
contacts_files.append(input_folder + "chr" + chr + ".5MB.K562." +
params.conttype) for chr in chr_nums
]
params.contacts_reader.read_files(
contacts_files,
coeff_fname=input_folder + "coefficient." + cell_type +
".25000.txt",
max_cpus=params.max_cpus,
fill_empty_contacts=fill_empty_contacts,
maxdist=params.maxdist,
expected_binsize=25000)
write_all_chrms_in_file = False #set True if you want write training file consisting several chromosomes
fill_empty_contacts = True #set True if you want use all contacts in region, without empty contacts
logging.getLogger(__name__).debug("Using input folder " + input_folder)
# Read contacts data
genome = fastaReader(input_folder + "sequence/hg38/hg38.fa",
name="hg38",
useOnlyChromosomes=["chr3"])
genome = genome.read_data()
# print(genome)
# print(genome.data.keys())
now = datetime.datetime.now()
params.contacts_reader = hicReader(fname=input_folder +
"H1/4DNFI2TK7L2F.hic",
genome=genome,
binsize=1000)
# params.contacts_reader = hicReader(fname=input_folder + "H1/control.chr4.50KBhic", genome=genome, binsize=1000)
params.contacts_reader = params.contacts_reader.read_data()
if params.use_only_contacts_with_CTCF == "cont_with_CTCF":
params.proportion = 1
params.contacts_reader.use_contacts_with_CTCF(
CTCFfile=input_folder +
"H1/CTCF/CTCF_H1_conservative_peaks.bed.gz",
maxdist=params.maxdist,
proportion=params.proportion,
keep_only_orient=params.keep_only_orient,
CTCForientfile=input_folder +
"H1/CTCF/CTCF_H1_conservative_peaks_orient.bed")
params.use_only_contacts_with_CTCF += str(