diff --git a/workflow/.DS_Store b/workflow/.DS_Store new file mode 100644 index 0000000000000000000000000000000000000000..a563988c14d65d2b69eb3b32e875d0f346a3055a Binary files /dev/null and b/workflow/.DS_Store differ diff --git a/workflow/main.nf b/workflow/main.nf index 29afb260c8dee81e663e4031fbf6625398eaa321..6c69edad2ea2dd0dc0af40c704c0e15ff974c932 100644 --- a/workflow/main.nf +++ b/workflow/main.nf @@ -198,7 +198,7 @@ process experimentQC { output: - file '*.{png,npz}' into deepToolsStats + file '*.{pdf,npz}' into deepToolsStats script: diff --git a/workflow/scripts/experiment_qc.py b/workflow/scripts/experiment_qc.py index 7386fcffd8e982d4d432e99becfc2aacb23ad2f2..34bce3a1516895044d32a9e8853b670b03255361 100644 --- a/workflow/scripts/experiment_qc.py +++ b/workflow/scripts/experiment_qc.py @@ -76,10 +76,10 @@ def generate_read_summary(design, extension): return mbs_filename -def check_correlation(mbs): +def check_spearman_correlation(mbs): '''Check Spearman pairwise correlation of samples based on read counts.''' - spearman_filename = 'heatmap_SpearmanCorr.png' + spearman_filename = 'heatmap_SpearmanCorr.pdf' spearman_params = "--corMethod spearman --skipZero" + \ " --plotTitle \"Spearman Correlation of Read Counts\"" + \ " --whatToPlot heatmap --colorMap RdYlBu --plotNumbers" @@ -95,12 +95,31 @@ def check_correlation(mbs): out, err = spearman_correlation.communicate() +def check_pearson_correlation(mbs): + '''Check Pearson pairwise correlation of samples based on read counts.''' + + pearson_filename = 'heatmap_PearsonCorr.pdf' + pearson_params = "--corMethod pearson --skipZero" + \ + " --plotTitle \"Pearson Correlation of Read Counts\"" + \ + " --whatToPlot heatmap --colorMap RdYlBu --plotNumbers" + + pearson_command = ( + "plotCorrelation -in %s %s -o %s" + % (mbs, pearson_params, pearson_filename) + ) + + logger.info("Running deeptools with %s", pearson_command) + + pearson_correlation = subprocess.Popen(pearson_command, shell=True) + out, err = pearson_correlation.communicate() + + def check_coverage(design, extension): '''Asses the sequencing depth of samples.''' bam_files = ' '.join(design['bam_reads']) labels = ' '.join(design['sample_id']) - coverage_filename = 'coverage.png' + coverage_filename = 'coverage.pdf' coverage_params = "-n 1000000 --plotTitle \"Sample Coverage\"" + \ " --ignoreDuplicates --minMappingQuality 10" @@ -136,7 +155,7 @@ def update_controls(design): def check_enrichment(sample_id, control_id, sample_reads, control_reads, extension): '''Asses the enrichment per sample.''' - fingerprint_filename = sample_id + '_fingerprint.png' + fingerprint_filename = sample_id + '_fingerprint.pdf' fingerprint_command = ( "plotFingerprint -b %s %s --extendReads %d --labels %s %s --plotFile %s" @@ -166,7 +185,9 @@ def main(): # Run correlation mbs_filename = generate_read_summary(design_df, extension) - check_correlation(mbs_filename) + check_spearman_correlation(mbs_filename) + check_pearson_correlation(mbs_filename) + # Run coverage check_coverage(design_df, extension) diff --git a/workflow/tests/test_experiment_qc.py b/workflow/tests/test_experiment_qc.py index 5256da5fbebc424ea29a1977d9257bfc95060ae3..3c4c9d0411faced4392b85f97e318d9ba58cf792 100644 --- a/workflow/tests/test_experiment_qc.py +++ b/workflow/tests/test_experiment_qc.py @@ -33,15 +33,17 @@ def test_check_update_controls(design_bam): @pytest.mark.singleend def test_experiment_qc_singleend(): assert os.path.exists(os.path.join(test_output_path, 'sample_mbs.npz')) - assert os.path.exists(os.path.join(test_output_path, 'heatmap_SpearmanCorr.png')) - assert os.path.exists(os.path.join(test_output_path, 'coverage.png')) - assert os.path.exists(os.path.join(test_output_path, 'ENCLB144FDT_fingerprint.png')) - assert os.path.exists(os.path.join(test_output_path, 'ENCLB831RUI_fingerprint.png')) + assert os.path.exists(os.path.join(test_output_path, 'heatmap_SpearmanCorr.pdf')) + assert os.path.exists(os.path.join(test_output_path, 'heatmap_PearsonCorr.pdf')) + assert os.path.exists(os.path.join(test_output_path, 'coverage.pdf')) + assert os.path.exists(os.path.join(test_output_path, 'ENCLB144FDT_fingerprint.pdf')) + assert os.path.exists(os.path.join(test_output_path, 'ENCLB831RUI_fingerprint.pdf')) @pytest.mark.pairdend def test_experiment_qc_pairedend(): assert os.path.exists(os.path.join(test_output_path, 'sample_mbs.npz')) - assert os.path.exists(os.path.join(test_output_path, 'heatmap_SpearmanCorr.png')) - assert os.path.exists(os.path.join(test_output_path, 'coverage.png')) - assert os.path.exists(os.path.join(test_output_path, 'ENCLB568IYX_fingerprint.png')) - assert os.path.exists(os.path.join(test_output_path, 'ENCLB637LZP_fingerprint.png')) + assert os.path.exists(os.path.join(test_output_path, 'heatmap_SpearmanCorr.pdf')) + assert os.path.exists(os.path.join(test_output_path, 'heatmap_PearsonCorr.pdf')) + assert os.path.exists(os.path.join(test_output_path, 'coverage.pdf')) + assert os.path.exists(os.path.join(test_output_path, 'ENCLB568IYX_fingerprint.pdf')) + assert os.path.exists(os.path.join(test_output_path, 'ENCLB637LZP_fingerprint.pdf'))