# chr_num="12"
# conttype = "contacts.gz"
if __name__ == '__main__': #Requered for parallization, at least on Windows
#,"chr10", "chr1"]:
for conttype in [conttype]:
print("hello")
logging.basicConfig(format='%(asctime)s %(name)s: %(message)s', datefmt='%I:%M:%S', level=logging.DEBUG)
input_folder ="/mnt/scratch/ws/psbelokopytova/202001051010polina_data/3DPredictor/input/K562/"
output_folder = "/mnt/scratch/ws/psbelokopytova/202001051010polina_data/3DPredictor/out/K562/5KB/all_predictors/"
cell_type="K562"
lengths_dict = {'chr1': 1494930, 'chr3': 609806, 'chr5': 518646, 'chr7': 682860, 'chr11': 726290, 'chr13': 115324}
params = Parameters()
params.binsize = 5000 #sequence resolution of contacts data. Use for finding of normalized coefficient file
params.window_size = params.binsize #region around contact to be binned for predictors. Usually equal to binsize
params.mindist = params.binsize*2+1 #minimum distance between contacting regions
params.maxdist = 1500000
params.sample_size = 250000 #how many contacts write to file
params.conttype = conttype
params.max_cpus = 11
params.keep_only_orient=False
params.use_only_contacts_with_CTCF = "all_cont"#"cont_with_CTCF"#"#"all_cont"#"cont_with_CTCF "
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
output_folder = args['output_folder']
cell_type = args['cell_type']
start = int(args['start'])
end = int(args['end'])
chromosome = 'chr' + args['chr_num']
hic_name = args['hic_name']
CTCF_file_name = args['CTCF_file_name']
#RNA_file_name = args['RNA_file_name']
# validate_chrs = args['validate_chrs'].split(",")
# for chr in validate_chrs:
# chr = int(chr)
params = Parameters()
params.binsize = int(
args['binsize']
) #sequence resolution of contacts data. Use for finding of normalized coefficient file
params.window_size = params.binsize #region around contact to be binned for predictors. Usually equal to binsize
params.mindist = params.binsize * 2 + 1 #minimum distance between contacting regions
params.maxdist = 1500000
# params.sample_size = end - start
params.sample_size = 2 #how many contacts write to file
#params.conttype = conttype
params.max_cpus = int(args['max_cpus'])
params.keep_only_orient = False
params.use_only_contacts_with_CTCF = "all_cont" #"all_cont" or "cont_with_CTCF"
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