def interaction_matrices_Chr(self):
"""
IntraChromosomal interaction matrices are produced by Homer at resolutions defined in the ini config file and plotted by HiCPlotter.
For more detailed information about the HOMER matrices visit: [HOMER matrices] (http://homer.ucsd.edu/homer/interactions/HiCmatrices.html)
For more detailed information about HiCPlotter visit: [HiCPlotter] (https://github.com/kcakdemir/HiCPlotter)
"""
jobs = []
chrs = config.param('interaction_matrices_Chr', 'chromosomes')
res_chr = config.param('interaction_matrices_Chr', 'resolution_chr').split(",")
if chrs == "All":
genome_dict = os.path.expandvars(config.param('DEFAULT', 'genome_dictionary', type='filepath'))
chrs = genome.chr_names_conv(genome_dict)
else:
chrs = chrs.split(",")
for sample in self.samples:
tagDirName = "_".join(("HTD", sample.name, self.enzyme))
homer_output_dir = os.path.join(self.output_dirs['homer_output_directory'], tagDirName)
sample_output_dir_chr = os.path.join(self.output_dirs['matrices_output_directory'], sample.name, "chromosomeMatrices")
# loop over chrs and res:
for res in res_chr:
for chr in chrs:
fileName = os.path.join(sample_output_dir_chr, "_".join((tagDirName, chr, res, "raw.txt")))
fileNameRN = os.path.join(sample_output_dir_chr, "_".join((tagDirName, chr, res, "rawRN.txt")))
fileNamePlot = os.path.join(sample_output_dir_chr, "".join((tagDirName,"_", chr,"_", res, "_raw-", chr, "\'.ofBins(0-\'*\')\'.", str(int(res)/1000), "K.jpeg")))
newFileNamePlot = os.path.join(sample_output_dir_chr, "".join((tagDirName,"_", chr,"_", res, "_raw-", chr, ".all.", str(int(res)/1000), "K.jpeg")))
jobMatrix = homer.hic_interactionMatrix_chr(sample.name, sample_output_dir_chr, homer_output_dir, res, chr, fileName, fileNameRN)
jobMatrix.samples = [sample]
jobPlot = hicplotter.intra_chrom_matrix_plot(
fileNameRN,
sample.name,
chr,
res,
os.path.join(sample_output_dir_chr, "_".join((tagDirName, chr, res, "raw"))),
fileNamePlot,
newFileNamePlot
)
jobPlot.samples = [sample]
jobs.extend([jobMatrix, jobPlot])
return jobs
def identify_TADs_TopDom(self):
"""
Topological associating Domains (TADs) are identified using TopDom at resolutions defined in the ini config file.
For more detailed information about the TopDom visit: [TopDom] (https://www.ncbi.nlm.nih.gov/pubmed/26704975)
"""
jobs = []
chrs = config.param('identify_TADs', 'chromosomes')
res_chr = config.param('identify_TADs', 'resolution_TADs').split(",")
if chrs == "All":
genome_dict = os.path.expandvars(config.param('DEFAULT', 'genome_dictionary', type='filepath'))
chrs = genome.chr_names_conv(genome_dict)
else:
chrs = chrs.split(",")
for sample in self.samples:
sample_output_dir = os.path.join(self.output_dirs['TAD_output_directory'], sample.name, "TopDom")
for res in res_chr:
for chr in chrs:
input_matrix = os.path.join(self.output_dirs['matrices_output_directory'], sample.name, "chromosomeMatrices", "_".join(("HTD", sample.name, self.enzyme, chr, res, "rawRN.txt")))
tmp_matrix = input_matrix + ".MatA"
output_matrix = os.path.join(sample_output_dir, "_".join(("HTD", sample.name, self.enzyme, chr, res, "rawRN.MatA.TopDom")))
output_script = "identify_TADs_TopDom." + sample.name + "_" + chr + "_res" + res + ".R"
job_inputFile = concat_jobs(
[
Job(command="mkdir -p " + sample_output_dir),
topdom.create_input(input_matrix, tmp_matrix, output_matrix, output_script, res)
],
name="identify_TADs.TopDom_create_input." + sample.name + "_" + chr + "_res" + res,
samples=[sample]
)
job_TADs = topdom.call_TADs(tmp_matrix, output_matrix, output_script)
job_TADs.name = "identify_TADs.TopDom_call_TADs." + sample.name + "_" + chr + "_res" + res
job_TADs.samples = [sample]
jobs.extend([
job_inputFile,
job_TADs
])
return jobs
def identify_TADs_RobusTAD(self):
"""
Topological associating Domain (TAD) scores are calculated using RobusTAD for every bin in the genome.
RobusTAD is resolution-independant and will use the first resolution in "resolution_TADs" under [identify_TADs] in the ini file.
For more detailed information about the RobusTAD visit: [RobusTAD] (https://github.com/rdali/RobusTAD)
"""
jobs = []
chrs = config.param('identify_TADs', 'chromosomes')
res = config.param('identify_TADs', 'resolution_TADs').split(",")[0]
if chrs == "All":
genome_dict = os.path.expandvars(config.param('DEFAULT', 'genome_dictionary', type='filepath'))
chrs = genome.chr_names_conv(genome_dict)
else:
chrs = chrs.split(",")
for sample in self.samples:
sample_output_dir = os.path.join(self.output_dirs['TAD_output_directory'], sample.name, "RobusTAD")
for chr in chrs:
input_matrix = os.path.join(self.output_dirs['matrices_output_directory'], sample.name, "chromosomeMatrices", "_".join(("HTD", sample.name, self.enzyme, chr, res, "rawRN.txt")))
prefix = os.path.splitext(os.path.basename(input_matrix))[0]
output_Scores = os.path.join(sample_output_dir, "".join(("BoundaryScores_", prefix, "_binSize" , str(int(res)/1000) ,"_minW250_maxW500_minRatio1.5.txt")))
output_calls = os.path.join(sample_output_dir, "".join(("TADBoundaryCalls_", prefix, "_binSize" , str(int(res)/1000) ,"_minW250_maxW500_minRatio1.5_threshold0.2.txt")))
job = concat_jobs([
Job(command="mkdir -p " + sample_output_dir),
robustad.call_TADs(input_matrix, sample_output_dir, res)
])
job.name = "identify_TADs.RobusTAD." + sample.name + "_" + chr
job.samples = [sample]
jobs.append(job)
return jobs
def identify_TADs(self):
"""
Topological associating Domains (TADs) are identified using TopDom at resolutions defined in the ini config file.
For more detailed information about the TopDom visit: [TopDom] (https://www.ncbi.nlm.nih.gov/pubmed/26704975)
"""
jobs = []
chrs = config.param('identify_TADs', 'chromosomes')
res_chr = config.param('identify_TADs', 'resolution_TADs').split(",")
if chrs == "All":
genome_dict = os.path.expandvars(
config.param('DEFAULT', 'genome_dictionary', type='filepath'))
chrs = genome.chr_names_conv(genome_dict)
else:
chrs = chrs.split(",")
for sample in self.samples:
sample_output_dir = os.path.join(
self.output_dirs['TAD_output_directory'], sample.name)
for res in res_chr:
for chr in chrs:
input_matrix = os.path.join(
self.output_dirs['matrices_output_directory'],
sample.name, "chromosomeMatrices", "_".join(
("HTD", sample.name, self.enzyme, chr, res,
"rawRN.txt")))
tmp_matrix = input_matrix + ".MatA"
output_matrix = os.path.join(
sample_output_dir, "_".join(
("HTD", sample.name, self.enzyme, chr, res,
"rawRN.MatA.TopDom")))
## make TopDom R script:
FileContent = """source(\\\"{script}\\\"); TopDom(matrix.file=\'{tmp_matrix}\', window.size={n}, outFile=\'{output_matrix}\')""".format(
script=os.path.expandvars(
"${R_TOOLS}/TopDom_v0.0.2.R"),
tmp_matrix=tmp_matrix,
n=config.param('identify_TADs', 'TopDom_n'),
output_matrix=output_matrix)
fileName = "identify_TADs_TopDom." + sample.name + "_" + chr + "_res" + res + ".R"
command_RFile = """echo \"{FileContent}\" > {fileName}""".format(
FileContent=FileContent, fileName=fileName)
command_TopDom = """mkdir -p {sample_output_dir} && {script} {input} {res}""".format(
sample_output_dir=sample_output_dir,
script="CreateTopDomMat.sh",
input=input_matrix,
res=res)
job_inputFile = Job(
input_files=[input_matrix],
output_files=[tmp_matrix],
module_entries=[["identify_TADs", "module_R"],
[
"identify_TADs",
"module_mugqic_tools"
]],
name="identify_TADs.create_input." + sample.name +
"_" + chr + "_res" + res,
command=command_RFile + " && " + command_TopDom,
removable_files=[tmp_matrix])
job_TADs = Job(
input_files=[tmp_matrix],
output_files=[
output_matrix + ".bed",
output_matrix + ".binSignal",
output_matrix + ".domain"
],
module_entries=[["identify_TADs", "module_R"],
[
"identify_TADs",
"module_mugqic_tools"
]],
name="identify_TADs.call_TADs." + sample.name + "_" +
chr + "_res" + res,
command="Rscript {fileName} && rm {fileName}".format(
fileName=fileName, tmp_matrix=tmp_matrix),
removable_files=[tmp_matrix])
jobs.extend([job_inputFile, job_TADs])
return jobs
def interaction_matrices_Chr(self):
"""
IntraChromosomal interaction matrices are produced by Homer at resolutions defined in the ini config file and plotted by HiCPlotter.
For more detailed information about the HOMER matrices visit: [HOMER matrices] (http://homer.ucsd.edu/homer/interactions/HiCmatrices.html)
For more detailed information about HiCPlotter visit: [HiCPlotter] (https://github.com/kcakdemir/HiCPlotter)
"""
jobs = []
chrs = config.param('interaction_matrices_Chr', 'chromosomes')
res_chr = config.param('interaction_matrices_Chr',
'resolution_chr').split(",")
if chrs == "All":
genome_dict = os.path.expandvars(
config.param('DEFAULT', 'genome_dictionary', type='filepath'))
chrs = genome.chr_names_conv(genome_dict)
else:
chrs = chrs.split(",")
for sample in self.samples:
tagDirName = "_".join(("HTD", sample.name, self.enzyme))
homer_output_dir = os.path.join(
self.output_dirs['homer_output_directory'], tagDirName)
sample_output_dir_chr = os.path.join(
self.output_dirs['matrices_output_directory'], sample.name,
"chromosomeMatrices")
# loop over chrs and res:
for res in res_chr:
for chr in chrs:
fileName = os.path.join(
sample_output_dir_chr, "_".join(
(tagDirName, chr, res, "raw.txt")))
fileNameRN = os.path.join(
sample_output_dir_chr, "_".join(
(tagDirName, chr, res, "rawRN.txt")))
jobMatrix = homer.hic_interactionMatrix_chr(
sample.name, sample_output_dir_chr, homer_output_dir,
res, chr, fileName, fileNameRN)
fileNamePlot = os.path.join(
sample_output_dir_chr,
"".join((tagDirName, "_", chr, "_", res, "_raw-",
chr, "\'.ofBins(0-\'*\')\'.",
str(int(res) / 1000), "K.jpeg")))
newFileNamePlot = os.path.join(
sample_output_dir_chr, "".join(
(tagDirName, "_", chr, "_", res, "_raw-", chr,
".all.", str(int(res) / 1000), "K.jpeg")))
commandChrPlot = "HiCPlotter.py -f {fileNameRN} -n {name} -chr {chr} -r {res} -fh 0 -o {sample_output_dir_chr} -ptr 0 -hmc {hmc} && mv {fileNamePlot} {newFileNamePlot}".format(
res=res,
chr=chr,
fileNameRN=fileNameRN,
name=sample.name,
sample_output_dir_chr=os.path.join(
sample_output_dir_chr, "_".join(
(tagDirName, chr, res, "raw"))),
hmc=config.param('interaction_matrices_Chr', 'hmc'),
fileNamePlot=fileNamePlot,
newFileNamePlot=newFileNamePlot)
jobPlot = Job(input_files=[fileNameRN],
output_files=[newFileNamePlot],
module_entries=[[
"interaction_matrices_Chr",
"module_HiCPlotter"
]],
name="interaction_matrices_Chr.plotting." +
sample.name + "_" + chr + "_res" + res,
command=commandChrPlot)
jobs.extend([jobMatrix, jobPlot])
return jobs