def test_hicConvertFormat_h5_to_homer():
outfile = NamedTemporaryFile(suffix='.homer', delete=False)
outfile.close()
args = "--matrices {} --outFileName {} --inputFormat cool --outputFormat homer ".format(
original_matrix_cool_chr4, outfile.name).split()
# hicConvertFormat.main(args)
compute(hicConvertFormat.main, args, 5)
test = hm.hiCMatrix(original_matrix_cool_chr4)
f = gzip.open(outfile.name, 'rb')
file_content = f.read()
outfile2 = NamedTemporaryFile(suffix='.homer', delete=False)
outfile2.close()
with open(outfile2.name, 'wb') as matrix_file:
matrix_file.write(file_content)
matrixFileHandlerInput = MatrixFileHandler(pFileType='homer',
pMatrixFile=outfile2.name)
_matrix, cut_intervals, nan_bins, \
distance_counts, correction_factors = matrixFileHandlerInput.load()
nt.assert_array_almost_equal(test.matrix.data, _matrix.data, decimal=0)
def test_plotSVL():
plot = NamedTemporaryFile(suffix='.png', delete=False)
outputFileName = NamedTemporaryFile(suffix='.txt', delete=False)
outputFileNameData = NamedTemporaryFile(suffix='.txt', delete=False)
plot.close()
outputFileName.close()
outputFileNameData.close()
matrix = ROOT + "small_test_matrix_50kb_res.h5"
matrix2 = ROOT + "small_test_matrix_50kb_res.h5"
args = "--matrices {} {} --plotFileName {} --outFileName {} --outFileNameData {} --dpi 300"\
.format(matrix, matrix2, plot.name, outputFileName.name, outputFileNameData.name).split()
# hicPlotSVL.main(args)
compute(hicPlotSVL.main, args, 5)
log.debug('--plotFileName {} --outFileName {} --outFileNameData {} '.format(plot.name, outputFileName.name, outputFileNameData.name))
log.debug('matrix {} {}'.format(matrix, matrix2))
log.debug('matrix {} {}'.format(matrix.split('/')[-1], matrix2.split('/')[-1]))
assert are_files_equal(ROOT + 'hicPlotSVL/data.txt', outputFileNameData.name, delta=2)
assert are_files_equal(ROOT + 'hicPlotSVL/p_values.txt', outputFileName.name, delta=2)
# res = compare_images(ROOT + 'hicPlotSVL/plot.png', plot.name, tol=50)
# assert res is None, res
os.unlink(plot.name)
os.unlink(outputFileName.name)
os.unlink(outputFileNameData.name)
def test_trivial_run(
matrices,
outputFormat,
resolutions,
):
"""
Test for all commandline arguments.
Options for cool input format are testet seperately.
"""
from pathlib import Path
# get suffix of input matrix without the dot
inputFormat = Path(matrices).suffix[1:]
# create file corresponding to output format
outFileName = NamedTemporaryFile(suffix=".{}".format(outputFormat),
delete=True)
args = "--matrices {} --outFileName {} --inputFormat {} --outputFormat {} {}".format(
matrices,
outFileName.name,
inputFormat,
outputFormat,
resolutions,
).split()
# hicConvertFormat.main(args)
compute(hicConvertFormat.main, args, 5)
def test_find_TADs_fdr():
# full test case with build of the matrix and search for tads
matrix = ROOT + "small_test_matrix.h5"
tad_folder = mkdtemp(prefix="test_case_find_tads_fdr")
args = "--matrix {} --minDepth 60000 --maxDepth 180000 --numberOfProcessors 2 --step 20000 \
--outPrefix {}/test_multiFDR --minBoundaryDistance 20000 \
--correctForMultipleTesting fdr --thresholdComparisons 0.1".format(
matrix, tad_folder).split()
# hicFindTADs.main(args)
compute(hicFindTADs.main, args, 5)
new = hm.hiCMatrix(tad_folder + "/test_multiFDR_zscore_matrix.h5")
test = hm.hiCMatrix(ROOT + 'find_TADs/FDR/multiFDR_zscore_matrix.h5')
nt.assert_equal(test.matrix.data, new.matrix.data)
nt.assert_equal(test.cut_intervals, new.cut_intervals)
print(tad_folder + "/test_multiFDR_boundaries.bed")
assert are_files_equal(ROOT + "find_TADs/FDR/multiFDR_boundaries.bed",
tad_folder + "/test_multiFDR_boundaries.bed")
assert are_files_equal(ROOT + "find_TADs/FDR/multiFDR_domains.bed",
tad_folder + "/test_multiFDR_domains.bed")
assert are_files_equal(ROOT + "find_TADs/FDR/multiFDR_tad_score.bm",
tad_folder + "/test_multiFDR_tad_score.bm")
assert are_files_equal(ROOT + "find_TADs/FDR/multiFDR_boundaries.gff",
tad_folder + "/test_multiFDR_boundaries.gff")
# assert are_files_equal
assert are_files_equal(ROOT + "find_TADs/FDR/multiFDR_score.bedgraph",
tad_folder + "/test_multiFDR_score.bedgraph")
shutil.rmtree(tad_folder)
def test_find_TADs_none():
# reduced test case, the z-score matrix is given to decrease run time
matrix = ROOT + "small_test_matrix.h5"
tad_folder = mkdtemp(prefix="test_case_find_tads_none")
shutil.copy(ROOT + "find_TADs/None/multiNone_tad_score.bm",
tad_folder + "/test_multiNone_tad_score.bm")
shutil.copy(ROOT + 'find_TADs/None/multiNone_zscore_matrix.h5',
tad_folder + "/test_multiNone_zscore_matrix.h5")
args = "--matrix {} --minDepth 60000 --maxDepth 180000 --numberOfProcessors 2 --step 20000 \
--outPrefix {}/test_multiNone --minBoundaryDistance 20000 \
--correctForMultipleTesting None --thresholdComparisons 1.0".format(
matrix, tad_folder).split()
# hicFindTADs.main(args)
compute(hicFindTADs.main, args, 5)
assert are_files_equal(ROOT + "find_TADs/None/multiNone_boundaries.bed",
tad_folder + "/test_multiNone_boundaries.bed")
assert are_files_equal(ROOT + "find_TADs/None/multiNone_domains.bed",
tad_folder + "/test_multiNone_domains.bed")
assert are_files_equal(ROOT + "find_TADs/None/multiNone_boundaries.gff",
tad_folder + "/test_multiNone_boundaries.gff")
assert are_files_equal(ROOT + "find_TADs/None/multiNone_score.bedgraph",
tad_folder + "/test_multiNone_score.bedgraph")
shutil.rmtree(tad_folder)
def test_hicAggregateContacts_cooler():
outfile_aggregate_plots = NamedTemporaryFile(suffix='.png',
prefix='hicaggregate_test_',
delete=False)
outfile_aggregate_matrix = NamedTemporaryFile(suffix='.tab',
prefix='hicaggregate_test_',
delete=False)
args = "--matrix {root}/Li_et_al_2015.cool --BED {root}/hicAggregateContacts/test_regions.bed " \
"--outFileName {out_agg} --numberOfBins 30 --range 50000:900000 --disable_bbox_tight --dpi 100 "\
"--mode intra-chr --perChr --outFilePrefixMatrix {out_mat}".\
format(root=ROOT, out_agg=outfile_aggregate_plots.name, out_mat=outfile_aggregate_matrix.name)
test_image_agg = ROOT + 'hicAggregateContacts/master_aggregate_intra_perChr.png' # noqa: F841
test_matrix_agg = ROOT + 'hicAggregateContacts/master_aggregate_intra_perChr.tab'
# hicexplorer.hicAggregateContacts.main(args.split())
compute(hicexplorer.hicAggregateContacts.main, args.split(), 5)
res = compare_images(test_image_agg, outfile_aggregate_plots.name,
tolerance)
assert res is None, res
assert are_files_equal(test_matrix_agg, outfile_aggregate_matrix.name)
os.remove(outfile_aggregate_plots.name)
os.remove(outfile_aggregate_matrix.name)
def test_normalize_smallest_cool(capsys):
outfile_one = NamedTemporaryFile(suffix='.cool', delete=False)
outfile_one.close()
outfile_two = NamedTemporaryFile(suffix='.cool', delete=False)
outfile_two.close()
args = "--matrices {} {} --normalize smallest -o {} {}".format(
matrix_one_cool, matrix_two_cool, outfile_one.name,
outfile_two.name).split()
# hicNormalize.main(args)
compute(hicNormalize.main, args, 5)
test_one = hm.hiCMatrix(ROOT + "/smallest_one.cool")
test_two = hm.hiCMatrix(ROOT + "/smallest_two.cool")
new_one = hm.hiCMatrix(outfile_one.name)
new_two = hm.hiCMatrix(outfile_two.name)
nt.assert_equal(test_one.matrix.data, new_one.matrix.data)
nt.assert_equal(test_one.cut_intervals, new_one.cut_intervals)
nt.assert_equal(test_two.matrix.data, new_two.matrix.data)
nt.assert_equal(test_two.cut_intervals, new_two.cut_intervals)
os.unlink(outfile_one.name)
os.unlink(outfile_two.name)
def test_build_matrix_restrictionCutFile_two(
sam1, sam2, outFile, qcFolder, outBam, binSize, restrictionCutFile,
minDistance, maxDistance, maxLibraryInsertSize, restrictionSequence,
danglingSequence, region, removeSelfLigation, minMappingQuality,
threads, inputBufferSize):
# test more args for restrictionCutFile option
region = genomicRegion(region)
args = "-s {} {} --restrictionCutFile {} --outFileName {} --QCfolder {} " \
"--restrictionSequence {} " \
"--danglingSequence {} " \
"--minDistance {} " \
"--maxLibraryInsertSize {} --threads {} " \
"--region {} --removeSelfLigation {} ".format(bam_R1, bam_R2,
restrictionCutFile, outFile.name,
qcFolder,
restrictionSequence,
danglingSequence,
minDistance,
maxLibraryInsertSize,
threads, region,
removeSelfLigation).split()
# hicBuildMatrix.main(args)
compute(hicBuildMatrix.main, args, 5)
os.unlink(outFile.name)
shutil.rmtree(qcFolder)
def test_hicConvertFormat_h5_to_ginteractions():
outfile = NamedTemporaryFile(suffix='.ginteractions', delete=False)
outfile.close()
args = "--matrices {} --outFileName {} --inputFormat h5 --outputFormat ginteractions ".format(
original_matrix_h5, outfile.name).split()
compute(hicConvertFormat.main, args, 5)
def test_build_matrix_chromosome_sizes(capsys):
outfile = NamedTemporaryFile(suffix='.h5', delete=False)
outfile.close()
outfile_bam = NamedTemporaryFile(suffix='.bam', delete=False)
outfile.close()
qc_folder = mkdtemp(prefix="testQC_")
args = "-s {} {} --outFileName {} -bs 5000 -b {} --QCfolder {} --threads 4 --chromosomeSizes {} \
--restrictionSequence GATC --danglingSequence GATC -rs {}".format(
sam_R1, sam_R2, outfile.name, outfile_bam.name, qc_folder,
ROOT + 'hicBuildMatrix/dm3.chrom.sizes', dpnii_file).split()
# hicBuildMatrix.main(args)
compute(hicBuildMatrix.main, args, 5)
test = hm.hiCMatrix(
ROOT + "hicBuildMatrix/chromosome_sizes/small_test_chromosome_size.h5")
new = hm.hiCMatrix(outfile.name)
nt.assert_equal(test.matrix.data, new.matrix.data)
nt.assert_equal(test.cut_intervals, new.cut_intervals)
# print("MATRIX NAME:", outfile.name)
print(set(os.listdir(ROOT + "QC/")))
assert are_files_equal(ROOT + "QC/QC.log", qc_folder + "/QC.log")
assert set(os.listdir(ROOT + "QC/")) == set(os.listdir(qc_folder))
# accept delta of 60 kb, file size is around 4.5 MB
assert abs(
os.path.getsize(ROOT + "hicBuildMatrix/chromosome_sizes/test.bam") -
os.path.getsize(outfile_bam.name)) < 64000
os.unlink(outfile.name)
shutil.rmtree(qc_folder)
def test_plot_region_interaction_file():
outfile = NamedTemporaryFile(suffix='.png',
prefix='viewpoint4',
delete=False)
outfile_interactions = NamedTemporaryFile(suffix='.bedgraph',
prefix='viewpoint_interactons',
delete=False)
matrix = ROOT + 'Li_et_al_2015.h5'
args = "--matrix {} --region X:3000000-3500000 -rp X:3200000-3300000 --outFileName {} -i {} --dpi 300".format(
matrix, outfile.name, outfile_interactions.name).split()
# hicPlotViewpoint.main(args)
compute(hicPlotViewpoint.main, args, 5)
res = compare_images(ROOT + 'hicPlotViewpoint/li_viewpoint_32-33Mb.png',
outfile.name,
tol=40)
assert res is None, res
assert are_files_equal(
ROOT + 'hicPlotViewpoint/li_32-33mb_interactions.bedgraph',
outfile_interactions.name)
os.remove(outfile.name)
os.remove(outfile_interactions.name)
def test_build_matrix_restrictionCutFile_eight(sam1, sam2, outFile, qcFolder, outBam, binSize,
restrictionCutFile, minDistance, maxDistance,
maxLibraryInsertSize, restrictionSequence,
danglingSequence, region, removeSelfLigation,
minMappingQuality, threads, inputBufferSize):
# added skipDuplicationCheck
args = "-s {} {} --restrictionCutFile {} --outFileName {} --QCfolder {} " \
"--restrictionSequence {} " \
"--danglingSequence {} " \
"--minDistance {} " \
"--maxLibraryInsertSize {} --threads {} " \
"--removeSelfLigation {} --keepSelfCircles " \
"--minMappingQuality {} --inputBufferSize {} " \
"--doTestRun --skipDuplicationCheck ".format(bam_R1, bam_R2,
restrictionCutFile, outFile.name,
qcFolder, restrictionSequence,
danglingSequence, minDistance,
maxLibraryInsertSize, threads,
removeSelfLigation,
minMappingQuality,
inputBufferSize).split()
# hicBuildMatrix.main(args)
compute(hicBuildMatrix.main, args, 5)
# os.unlink(outFile.name)
shutil.rmtree(qcFolder)
def test_main_one_file():
outfile_domains = NamedTemporaryFile(suffix='.txt', delete=True)
args = "-d {} -om {} ".format(
ROOT + 'hicMergeDomains/10kbtad_domains.bed',
outfile_domains.name).split()
# hicMergeDomains.main(args)
compute(hicMergeDomains.main, args, 5)
def test_build_matrix_restriction_enzyme(capsys):
outfile = NamedTemporaryFile(suffix='.h5', delete=False)
outfile.close()
outfile_bam = NamedTemporaryFile(suffix='.bam', delete=False)
outfile.close()
qc_folder = mkdtemp(prefix="testQC_")
args = "-s {} {} --outFileName {} -bs 5000 -b {} --QCfolder {} --threads 4 --danglingSequence GATC AGCT --restrictionSequence GATC AAGCTT -rs {} {}".format(
sam_R1, sam_R2, outfile.name, outfile_bam.name, qc_folder, dpnii_file,
ROOT + 'hicFindRestSite/hindIII.bed').split()
# hicBuildMatrix.main(args)
compute(hicBuildMatrix.main, args, 5)
test = hm.hiCMatrix(ROOT + "small_test_matrix_parallel_two_rc.h5")
new = hm.hiCMatrix(outfile.name)
nt.assert_equal(test.matrix.data, new.matrix.data)
nt.assert_equal(test.cut_intervals, new.cut_intervals)
# print("MATRIX NAME:", outfile.name)
print(set(os.listdir(ROOT + "QC_multi_restriction/")))
assert are_files_equal(ROOT + "QC_multi_restriction/QC.log",
qc_folder + "/QC.log")
assert set(os.listdir(ROOT + "QC_multi_restriction/")) == set(
os.listdir(qc_folder))
# accept delta of 60 kb, file size is around 4.5 MB
assert abs(
os.path.getsize(ROOT + "small_test_matrix_result.bam") -
os.path.getsize(outfile_bam.name)) < 64000
os.unlink(outfile.name)
shutil.rmtree(qc_folder)
def test_trivial_run(matrices, labels, plotsize1, plotsize2, perchr,
skipDiagonal, maxdepth, chromosomeExclude):
"""
Simple test for general behaviour with all commandline args.
"""
outfile_new = NamedTemporaryFile(suffix='.png',
prefix='plotFile',
delete=False)
outFileData_new = NamedTemporaryFile(suffix='.txt',
prefix='dataFile',
delete=False)
args = "--matrices {} --plotFile {} {} --plotsize {} {} {} --outFileData {} --maxdepth {} --chromosomeExclude {}".format(
matrices,
outfile_new.name,
labels,
plotsize1,
plotsize2,
perchr,
outFileData_new.name,
maxdepth,
chromosomeExclude,
).split()
# with open("test.txt", 'a') as fo:
# fo.write(str(args))
# fo.write("\n")
# hicPlotDistVsCounts.main(args)
compute(hicPlotDistVsCounts.main, args, 5)
def test_hicAggregateContacts_clustering_cool():
outfile_aggregate_plots = NamedTemporaryFile(suffix='.png',
prefix='hicaggregate_test_',
delete=False)
outfile_heatmaps = NamedTemporaryFile(suffix='.png',
prefix='hicaggregate_heatmap_',
delete=False)
args = "--matrix {root}/Li_et_al_2015.cool --BED {root}/hicAggregateContacts/test_regions.bed " \
"--outFileName {out_agg} --numberOfBins 30 --range 50000:900000 --hclust 4 " \
"--diagnosticHeatmapFile {out_heat} --howToCluster diagonal --disable_bbox_tight --dpi 100 " \
"--BED2 {root}/hicAggregateContacts/test_regions.bed".format(root=ROOT, out_agg=outfile_aggregate_plots.name,
out_heat=outfile_heatmaps.name)
test_image_agg = ROOT + 'hicAggregateContacts/master_aggregate_hclust4.png'
test_image_heatmap = ROOT + 'hicAggregateContacts/master_heatmap.png'
# hicexplorer.hicAggregateContacts.main(args.split())
compute(hicexplorer.hicAggregateContacts.main, args.split(), 5)
res = compare_images(test_image_agg, outfile_aggregate_plots.name,
tolerance)
assert res is None, res
res = compare_images(test_image_heatmap, outfile_heatmaps.name, tolerance)
assert res is None, res
os.remove(outfile_aggregate_plots.name)
os.remove(outfile_heatmaps.name)
def test_main_h5():
outfile_loop_h5 = NamedTemporaryFile(suffix='.bedgraph', delete=True)
args = "--matrix {} -o {} -pit 1 -p 0.5 -pp 0.5 -t 4 -tpc 4".format(
ROOT + "small_test_matrix.h5", outfile_loop_h5.name).split()
# hicDetectLoops.main(args)
compute(hicDetectLoops.main, args, 5)
def test_build_matrix_rf_multi():
outfile = NamedTemporaryFile(suffix='.h5', delete=False)
outfile.close()
qc_folder = mkdtemp(prefix="testQC_")
args = "-s {} {} -rs {} {} --outFileName {} --QCfolder {} " \
"--restrictionSequence GATC AAGCTT " \
"--danglingSequence GATC AGCT " \
"--minDistance 150 " \
"--maxLibraryInsertSize 1500 --threads 4".format(sam_R1, sam_R2, dpnii_file, ROOT + 'hicFindRestSite/hindIII.bed',
outfile.name,
qc_folder).split()
# --danglingSequence GATC AGCT --restrictionSequence GATC AAGCTT
# hicBuildMatrix.main(args)
compute(hicBuildMatrix.main, args, 5)
test = hm.hiCMatrix(ROOT + "small_test_rf_matrix_multiple.h5")
new = hm.hiCMatrix(outfile.name)
nt.assert_equal(test.matrix.data, new.matrix.data)
nt.assert_equal(test.cut_intervals, new.cut_intervals)
print(set(os.listdir(ROOT + "QC_rc_multiple/")))
assert are_files_equal(ROOT + "QC_rc_multiple/QC.log",
qc_folder + "/QC.log")
assert set(os.listdir(ROOT + "QC_rc_multiple/")) == set(
os.listdir(qc_folder))
os.unlink(outfile.name)
shutil.rmtree(qc_folder)
def test_build_matrix_cooler():
outfile = NamedTemporaryFile(suffix='.cool', delete=False)
outfile.close()
outfile_bam = NamedTemporaryFile(suffix='.bam', delete=False)
outfile.close()
qc_folder = mkdtemp(prefix="testQC_")
args = "-s {} {} --outFileName {} -bs 5000 -b {} --QCfolder {} --threads 4 \
--restrictionSequence GATC --danglingSequence GATC -rs {}".format(
sam_R1, sam_R2, outfile.name, outfile_bam.name, qc_folder,
dpnii_file).split()
# hicBuildMatrix.main(args)
compute(hicBuildMatrix.main, args, 5)
test = hm.hiCMatrix(ROOT + "small_test_matrix_parallel.h5")
new = hm.hiCMatrix(outfile.name)
nt.assert_equal(test.matrix.data, new.matrix.data)
# nt.assert_equal(test.cut_intervals, new.cut_intervals)
nt.assert_equal(len(new.cut_intervals), len(test.cut_intervals))
cut_interval_new_ = []
cut_interval_test_ = []
for x in new.cut_intervals:
cut_interval_new_.append(x[:3])
for x in test.cut_intervals:
cut_interval_test_.append(x[:3])
nt.assert_equal(cut_interval_new_, cut_interval_test_)
# print(set(os.listdir(ROOT + "QC/")))
assert are_files_equal(ROOT + "QC/QC.log", qc_folder + "/QC.log")
assert set(os.listdir(ROOT + "QC/")) == set(os.listdir(qc_folder))
os.unlink(outfile.name)
shutil.rmtree(qc_folder)
def test_plot_single_point_interaction_file_two_matrices():
outfile = NamedTemporaryFile(suffix='.png',
prefix='viewpoint2',
delete=False)
outfile_interactions_one = NamedTemporaryFile(
suffix='.bedgraph',
prefix='viewpoint_interactons_Li_et_al_2015.h5',
delete=False)
outfile_interactions_two = NamedTemporaryFile(
suffix='.bedgraph',
prefix='viewpoint_interactons_Li_et_al_2015_twice.h5',
delete=False)
matrix = ROOT + 'Li_et_al_2015.h5' + ' ' + ROOT + 'Li_et_al_2015_twice.h5'
args = "--matrix {} --region X:3000000-3500000 -rp X:3200000 --outFileName {} -i {} --dpi 300".format(
matrix, outfile.name, 'viewpoint_interactons').split()
# hicPlotViewpoint.main(args)
compute(hicPlotViewpoint.main, args, 5)
res = compare_images(ROOT +
'/hicPlotViewpoint/li_viewpoint_32Mb_twice.png',
outfile.name,
tol=40)
assert res is None, res
assert are_files_equal(
ROOT + '/hicPlotViewpoint/li_32mb_interactions_one.bedgraph',
outfile_interactions_one.name)
assert are_files_equal(
ROOT + '/hicPlotViewpoint/li_32mb_interactions_two.bedgraph',
outfile_interactions_two.name)
os.remove(outfile.name)
os.remove(outfile_interactions_one.name)
os.remove(outfile_interactions_two.name)
def test_hicConvertFormat_h5_to_hicpro():
outfile = NamedTemporaryFile(suffix='.hicpro', delete=False)
outfile_bed = NamedTemporaryFile(suffix='.bed', delete=False)
outfile.close()
args = "--matrices {} --outFileName {} --inputFormat h5 --outputFormat hicpro --bedFileHicpro {}".format(
original_matrix_h5, outfile.name, outfile_bed.name).split()
compute(hicConvertFormat.main, args, 5)
def test_cool_all_multi_core_all_multichr_chromosome_not_in_matrix():
outfile_pref = NamedTemporaryFile(prefix='differential_tad', delete=True)
args = "--targetMatrix {} --controlMatrix {} --tadDomains {} -t {} -o {} -m {} -mr {}".format(
ROOT + "GSM2644945_Untreated-R1.100000_chr1.cool",
ROOT + "GSM2644947_Auxin2days-R1.100000_chr1.cool",
ROOT + "untreated_R1_domains_chr1_chr2.bed", 4, outfile_pref.name,
'all', 'all').split()
compute(hicDifferentialTAD.main, args, 5)
def test_hicAggregateContacts_chromosome_not_given():
outfile_aggregate_plots = NamedTemporaryFile(suffix='.png', prefix='hicaggregate_test_', delete=False)
args = "--matrix {root}/Li_et_al_2015.h5 --BED {root}/hicAggregateContacts/test_regions_region_not_given.bed " \
"--outFileName {out_agg} --numberOfBins 30 --range 50000:900000 --disable_bbox_tight --dpi 100 "\
"--mode intra-chr --perChr ".format(root=ROOT, out_agg=outfile_aggregate_plots.name)
compute(hicexplorer.hicAggregateContacts.main, args.split(), 5)
def test_pca_bigwig_gene_density_intermediate_matrices_norm():
pca1 = NamedTemporaryFile(suffix='.bw', delete=False)
pca2 = NamedTemporaryFile(suffix='.bw', delete=False)
pearson_matrix = NamedTemporaryFile(suffix='.cool', delete=False)
obs_exp_matrix = NamedTemporaryFile(suffix='.cool', delete=False)
pca1.close()
pca2.close()
pearson_matrix.close()
obs_exp_matrix.close()
matrix = ROOT + "small_test_matrix.h5"
gene_track = ROOT + 'dm3_genes.bed.gz'
chromosomes = 'chrX chrXHet'
args = "--matrix {} --outputFileName {} {} -f bigwig -noe 2 \
--extraTrack {} --chromosomes {} --pearsonMatrix {} --obsexpMatrix {} \
--method dist_norm --ligation_factor"\
.format(matrix, pca1.name, pca2.name, gene_track, chromosomes,
pearson_matrix.name, obs_exp_matrix.name).split()
# hicPCA.main(args)
compute(hicPCA.main, args, 5)
chrom_list = ['chrX', 'chrXHet']
assert are_files_equal_bigwig(ROOT + "hicPCA/pca1_gene_track.bw",
pca1.name, chrom_list)
assert are_files_equal_bigwig(ROOT + "hicPCA/pca2_gene_track.bw",
pca2.name, chrom_list)
test_pearson = hm.hiCMatrix(ROOT + "hicPCA/pearson_norm.cool")
new_pearson = hm.hiCMatrix(pearson_matrix.name)
test_obs_exp = hm.hiCMatrix(ROOT + "hicPCA/obsexp_norm.cool")
new_obs_exp = hm.hiCMatrix(obs_exp_matrix.name)
# load h5 matrices to compare if they are store the same data
test_obs_exp_h5 = hm.hiCMatrix(ROOT + "hicPCA/obsexp_norm.h5")
test_pearson_h5 = hm.hiCMatrix(ROOT + "hicPCA/pearson_norm.h5")
nt.assert_array_almost_equal(test_pearson.matrix.data,
new_pearson.matrix.data,
decimal=DELTA_DECIMAL)
nt.assert_array_almost_equal(test_obs_exp.matrix.data,
new_obs_exp.matrix.data,
decimal=DELTA_DECIMAL)
nt.assert_array_almost_equal(test_pearson_h5.matrix.data,
new_pearson.matrix.data,
decimal=DELTA_DECIMAL)
nt.assert_array_almost_equal(test_obs_exp_h5.matrix.data,
new_obs_exp.matrix.data,
decimal=DELTA_DECIMAL)
os.unlink(pca1.name)
os.unlink(pca2.name)
os.unlink(obs_exp_matrix.name)
os.unlink(pearson_matrix.name)
def test_main_one_file_protein():
outfile_domains = NamedTemporaryFile(suffix='.txt', delete=True)
args = "-d {} -om {} -p {} ".format(
ROOT + 'hicMergeDomains/10kbtad_domains.bed',
outfile_domains.name, ROOT + 'hicMergeDomains/ctcf_sorted.bed').split()
# hicMergeDomains.main(args)
compute(hicMergeDomains.main, args, 5)
are_files_equal(outfile_domains.name, ROOT + 'hicMergeDomains/one_file', delta=2)
def test_build_matrix_fail(capsys):
outfile = NamedTemporaryFile(suffix='', delete=False)
outfile.close()
outfile_bam = NamedTemporaryFile(suffix='.bam', delete=False)
outfile.close()
qc_folder = mkdtemp(prefix="testQC_")
args = "-s {} {} --outFileName {} -bs 5000 -b {} --QCfolder {} --threads 4 ".format(
sam_R1, sam_R2, outfile_bam.name, outfile.name, qc_folder).split()
# hicBuildMatrix.main(args)
compute(hicBuildMatrix.main, args, 5)
def test_main_cool_chromosomes_threads_inner_threads():
outfile_loop_cool = NamedTemporaryFile(suffix='.bedgraph', delete=True)
args = "--matrix {} -o {} --maxLoopDistance 3000000 -pit 1 -w 5 -pw 2 -p 0.5 -pp 0.55 --chromosomes 1 2 --threads 2 --threadsPerChromosome 4".format(
ROOT + "hicDetectLoops/GSE63525_GM12878_insitu_primary_2_5mb.cool",
outfile_loop_cool.name).split()
# hicDetectLoops.main(args)
compute(hicDetectLoops.main, args, 5)
assert are_files_equal(ROOT + "hicDetectLoops/loops.bedgraph",
outfile_loop_cool.name,
delta=0)
def test_hicConvertFormat_h5_to_mcool():
outfile = NamedTemporaryFile(suffix='.cool', delete=False)
outfile.close()
args = "--matrices {} --outFileName {} --inputFormat h5 --outputFormat mcool -r 10000 100000 200000 ".format(
original_matrix_h5, outfile.name).split()
compute(hicConvertFormat.main, args, 5)
new1 = hm.hiCMatrix(outfile.name + '::/resolutions/10000') # noqa: F841
new2 = hm.hiCMatrix(outfile.name + '::/resolutions/100000') # noqa: F841
new3 = hm.hiCMatrix(outfile.name + '::/resolutions/200000') # noqa: F841
def test_compute_background_functional():
outfile = NamedTemporaryFile(suffix='.txt', delete=False)
outfile.close()
args = "--matrices {} {} --referencePoints {} -o {} -t {}".format(
ROOT + 'FL-E13-5_chr1.cool', ROOT + 'MB-E10-5_chr1.cool',
ROOT + 'referencePoints.bed', outfile.name, 1).split()
# chicViewpointBackgroundModel.main(args)
compute(chicViewpointBackgroundModel.main, args, 5)
assert are_files_equal(ROOT + 'background.txt',
outfile.name,
delta=700,
skip=1)
def test_trivial_run_xfail(matrix, outFileName, chromosomes, action):
"""
Test checks if all commandline args work in general.
"""
args = "--matrix {} --outFileName {} --chromosomes {} --action {}".format(
matrix,
outFileName.name,
chromosomes,
action,
).split()
# hicAdjustMatrix.main(args)
compute(hicAdjustMatrix.main, args, 5)
