hm_file = "/mnt/storage/home/vsfishman/HiC/tutorial_Fishman/chick/mapped-GalGal5filtered/GalGal5filteredChrmLevel/ChEF-all-HindIII-40k.hm.IC"
domains_files_Arm = "mapped-GalGal5filtered/GalGal5filteredChrmLevel/ChEF-all-HindIII-40k.hm.gzipped_matrix/ChEF-all-HindIII-40k.hm.gzipped_matrix.jucebox_domains.annotation"
domains_files_Dix = "/mnt/storage/home/vsfishman/HiC/data/chick/DixonDomainsChEF_all_HindIII_40k.hm.IC_domains_40KB/DixonDomainsChEF_all_HindIII_40k.hm.IC_domains_40KB.jucebox_domains.annotation"
second_hm_file = "/mnt/storage/home/vsfishman/HiC/tutorial_Fishman/chick/mapped-GalGal5filtered/GalGal5filteredChrmLevel/Blood-all-HindIII-40k.hm.IC"
second_domains_files_Arm = "mapped-GalGal5filtered/GalGal5filteredChrmLevel/Blood-all-HindIII-40k.hm.gzipped_matrix/Blood-all-HindIII-40k.hm.gzipped_matrix.jucebox_domains.annotation"
second_domains_files_Dix = "/mnt/storage/home/vsfishman/HiC/data/chick/DixonDomainsBlood_all_HindIII_40k.hm.IC_domains_40KB/DixonDomainsBlood_all_HindIII_40k.hm.IC_domains_40KB.jucebox_domains.annotation"
resolution = extractResolutionFromFileName(hm_file)
assert resolution == extractResolutionFromFileName(second_hm_file)
data_dict = {} # a structure to keep chrms arrays
data_dict_second_hmap = {} # a structure to keep chrms arrays
for chrm,array in enumerate(get_chromosomes(hm_file,genome_db_contig,resolution)):
data_dict[chrm] = array
for chrm,array in enumerate(get_chromosomes(second_hm_file,genome_db_contig,resolution)):
data_dict_second_hmap[chrm] = array
#chrm,nt_start,nt_end,title
#######################################ADD REGIONS TO PLOT HERE
###EXAMPLE:
###points = [
#["chr4",40100000,44000000,"Chromomer_Anja"],
#["chr5",4000,440000,"Chromomer_Anja"]
#]
points = [["chr15",7200000,7280000,"Gamatu_Gene"]]
#points = [["chr15",6200000,8280000,"Gamatu_Gene"]]
np.savetxt(results_file_name + ".criteias.txt",
criteias,
header="distances(columns): " +
"\t".join(map(str, sorted(distances.keys()))) +
" percentiles(rows): " + "\t".join(map(str, percentiles)))
results_file = open(results_file_name, "w")
results_file.write("Contig\tNt\tChr\tNt\tMax_percentile\t" +
"\t".join(map(str, sorted(distances.keys()))) + "\n")
print "Calculating..."
result_counts = dict([(p, 0) for p in percentiles])
for chr_numb, data in enumerate(
get_chromosomes(hm_file, genome_db_contig, resolution)):
for bin in xrange(
max_distance + min_distance_from_chrm_end_in_bins,
len(data) - max_distance - min_distance_from_chrm_end_in_bins):
max_bin_percentile = 101 #percentile can not be > 100, so 101 indicates we are in the begining of the loop
sign = None #sign means left contacts > right contacts (True), left contacts < right contacts (False), or not defined (None)
ratios = []
for jnd, dist in enumerate(sorted(distances.keys())):
max_curren_percentile = 101
left = data[bin][bin - dist] #contacts on the left side of the bin
right = data[bin][bin +
dist] #contacts on the right side of the bin
if left * right == 0: #bin has 0 contacts on left or right side
max_bin_percentile = 101
break #than we do not consider it at all (max_percentile will be -1)
if len(points)==0:
print "No points relevant for threshold found in a file",results_file
continue
all_result_files_points[results_file] = points
figs_dir = results_file+".tr_"+str(threshold)+".figs/"
if os.path.exists(figs_dir):
import shutil
shutil.rmtree(figs_dir)
os.mkdir(figs_dir)
assert len(all_result_files_points) > 0
data_dict = {} # a structure to keep chrms arrays
if len(np.unique([p[0] for p in points for points in all_result_files_points.values()]))==1:
data_dict[points[0][0]] = get_chromosomes(hm_file,genome_db,resolution,chrNumb=points[0][0])
else:
for chrm,array in enumerate(get_chromosomes(hm_file,genome_db,resolution)):
data_dict[chrm] = array
for results_file in all_result_files_points:
points = all_result_files_points[results_file]
figs_dir = results_file+".tr_"+str(threshold)+".figs/"
for i in points:
chrm,nt = i
bin = int(nt/resolution)
data = data_dict[chrm]
#data is a matrix (2D numpu array)
left_border = max(0,bin-distance)
right_border = min(len(data),bin+distance)
to_plot = np.log(data[left_border:right_border,left_border:right_border])
import shutil
shutil.rmtree(final_figs_dir)
os.mkdir(final_figs_dir)
assert len(all_result_files_points) > 0
data_dict = {} # a structure to keep chrms arrays
data_dict_second_hmap = {} # a structure to keep chrms arrays
if len(
np.unique([
p[0] for p in points
for points in all_result_files_points.values()
])) == 1:
data_dict[points[0][0]] = get_chromosomes(hm_file,
genome_db_contig,
resolution,
chrNumb=points[0][0])
data_dict_second_hmap[points[0][0]] = get_chromosomes(second_hm_file,
genome_db_contig,
resolution,
chrNumb=points[0][0])
else:
for chrm, array in enumerate(
get_chromosomes(hm_file, genome_db_contig, resolution)):
data_dict[chrm] = array
for chrm, array in enumerate(
get_chromosomes(second_hm_file, genome_db_contig, resolution)):
data_dict_second_hmap[chrm] = array
final_results = open(hm_file + ".FISH_regions.txt", "w")
for results_file in all_result_files_points:
np.savetxt(results_file_name + ".criteias.txt",
criteias,
header="distances(columns): " +
"\t".join(map(str, sorted(distances.keys()))) +
" percentiles(rows): " + "\t".join(map(str, percentiles)))
results_file = open(results_file_name, "w")
results_file.write("chr\tNucleotide\tMax_percentile\t" +
"\t".join(map(str, sorted(distances.keys()))) + "\n")
print "Calculating..."
result_counts = dict([(p, 0) for p in percentiles])
for chr_numb, data in enumerate(get_chromosomes(hm_file, genome_db,
resolution)):
for bin in xrange(
max_distance + min_distance_from_chrm_end_in_bins,
len(data) - max_distance - min_distance_from_chrm_end_in_bins):
max_bin_percentile = 101 #percentile can not be > 100, so 101 indicates we are in the begining of the loop
sign = None #sign means left contacts > right contacts (True), left contacts < right contacts (False), or not defined (None)
ratios = []
for jnd, dist in enumerate(sorted(distances.keys())):
max_curren_percentile = 101
left = data[bin][bin - dist] #contacts on the left side of the bin
right = data[bin][bin +
dist] #contacts on the right side of the bin
if left * right == 0: #bin has 0 contacts on left or right side
max_bin_percentile = 101
break #than we do not consider it at all (max_percentile will be -1)
genome_db = Genome("/mnt/storage/home/vsfishman/HiC/fasta/" + genomeName,
readChrms=["#", "X", "Y"])
#hm_file = "/mnt/storage/home/vsfishman/HiC/data/mESC/mapped-mm10/mm10/mESC-all-HindIII_refined.frag_res25k_hiRes.hm.IC.cis.hdf5"
hm_file = "/mnt/storage/home/vsfishman/HiC/data/mESC/mapped-mm10/mm10/mESC-all-HindIII_refined.frag_res25k_hiRes.hm.IC.cis.hdf5"
#hm_file = "/mnt/storage/home/vsfishman/HiC/data/mESC/mapped-mm10/mm10/mESC-all-HindIII_refined.frag_res10k_hiRes.hm.IC.cis.hdf5"
resolution = extractResolutionFromFileName(hm_file)
distances = range(2, 11) + [30, 50]
min_distance_from_chrm_end_in_bins = 100000 / resolution
result = {}
for d in distances:
result[d] = []
for data in get_chromosomes(hm_file, genome_db, resolution):
for dist in distances:
for bin in xrange(
dist + min_distance_from_chrm_end_in_bins,
len(data) - dist - min_distance_from_chrm_end_in_bins):
left = data[bin][bin - dist]
right = data[bin][bin + dist]
if left * right == 0:
continue
ratio = max(left, right) / min(left, right)
result[dist].append(ratio)
print "min\tmax\tmean\tmedian\n"
if os.path.exists(hm_file + ".stat"):
import shutil
shutil.rmtree(hm_file + ".stat")
from mirnylib import h5dict
from mirnylib import plotting
from hiclib import binnedData
from hiclib import fragmentHiC
import math
genome_db = genome.Genome(
"/mnt/storage/home/vsfishman/HiC/fasta/GalGal5/GCF_000002315.4_Gallus_gallus-5.0_assembly_structure/Primary_Assembly/assembled_chromosomes/FASTA/",
readChrms=[],
chrmFileTemplate="%s.fna")
hm_file = sys.argv[1]
figure_path = "/mnt/storage/home/vsfishman/HiC/pics/"
domain_res = extractResolutionFromFileName(hm_file)
all_chrms = get_chromosomes(hm_file, genome_db, domain_res)
chrms = range(genome_db.chrmCount) #numner of chrms in genome
st = genome_db.chrmStartsBinCont #array of numbers of chrms start (in bins)
end = genome_db.chrmEndsBinCont #array of numbers of chrms ends (in bins). Numer end[-1] is not in chromosome (this number is 1st bin of next chrm)
def plot_one_chr_fragment(matrix, figure_path, chr_name, ch_start, ch_end):
print "Plotting picture"
ch_start = ch_start / domain_res
ch_end = ch_end / domain_res
# domain_st = domain_st / domain_res
# domain_end = domain_end / domain_res
i = genome_db.label2idx[chr_name]
q2_0 = matrix[i]
q2_0 = q2_0[ch_start:ch_end, ch_start:ch_end]
domains=args.domains.split(",")
colors=args.colors.split(",")
else:
domains=[]
colors=[]
assert len(domains)==len(colors)
domains=[np.genfromtxt(domain_file,dtype=np.dtype([('chrm','S10'),('start',np.uint32),('end',np.uint32)]),usecols = (0,1,2))
for domain_file in domains]
resolution = extractResolutionFromFileName(hm_file)
print "Using resolution ",resolution
data_dict = {} # a structure to keep chrms arrays
data_dict_second_hmap = {} # a structure to keep chrms arrays
for chrm,array in enumerate(get_chromosomes(hm_file,genome_db,resolution)):
data_dict[chrm] = array
if args.reg == "all":
args.reg = ",".join([i+":0-"+str(genome_db.chrmLens[genome_db.label2idx[i]]) for i in genome_db.chrmLabels
if genome_db.chrmLens[genome_db.label2idx[i]] > resolution*10])
for i in args.reg.split(","):
chrm = i.split(":")[0]
nt_start,nt_end = map(int,i.split(":")[1].split("-"))
title = "_".join(map(str,[chrm,nt_start,nt_end]))
chrmLabel = chrm
chrm = genome_db.label2idx[chrm]
data = data_dict[chrm]
left_border = nt_start/resolution
right_border = nt_end/resolution