qc/pipeline: fix bug locating reference data if organism name contains spaces.

Updates the QC pipeline driver to use the normalised organism name (converted to lower case and whitespace converted to underscores), otherwise reference data cannot be located for names which include whitespace.

Author: pjbriggs

auto_process_ngs/qc/pipeline.py

@@ -452,7 +452,7 @@ def add_project(self,project,protocol,qc_dir=None,organism=None,
             get_star_index = GetReferenceDataset(
                 "%s: get STAR index for '%s'" % (project.name,
                                                  organism),
-                organism,
+                organism_name,
                 self.params.star_indexes,
                 force_reference=self.params.force_star_index)
             self.add_task(get_star_index)
@@ -480,7 +480,7 @@ def add_project(self,project,protocol,qc_dir=None,organism=None,
                 "%s: get RSeQC reference gene model for '%s'" %
                 (project.name,
                  organism),
-                organism,
+                organism_name,
                 self.params.annotation_bed_files)
             self.add_task(get_reference_gene_model)
             qc_metadata['annotation_bed'] = \
@@ -490,7 +490,7 @@ def add_project(self,project,protocol,qc_dir=None,organism=None,
             get_annotation_gtf = GetReferenceDataset(
                 "%s: get GTF annotation for '%s'" % (project.name,
                                                      organism),
-                organism,
+                organism_name,
                 self.params.annotation_gtf_files,
                 force_reference=self.params.force_gtf_annotation)
             self.add_task(get_annotation_gtf)

auto_process_ngs/test/qc/protocols/test_standard_qc.py

@@ -317,4 +317,91 @@ def test_qcpipeline_standard_se(self):
                   "multiqc_report.html"):
             self.assertTrue(os.path.exists(os.path.join(self.wd,
                                                         "PJB",f)),
-                            "Missing %s" % f)
+                            "Missing %s" % f)
+
+    def test_qcpipeline_standard_paired_end_organism_with_whitespace(self):
+        """QCPipeline: standard QC run (paired-end data, organism name contains whitespace)
+        """
+        # Make mock QC executables
+        MockFastqScreen.create(os.path.join(self.bin,"fastq_screen"))
+        MockFastQC.create(os.path.join(self.bin,"fastqc"))
+        MockStar.create(os.path.join(self.bin,"STAR"))
+        MockSamtools.create(os.path.join(self.bin,"samtools"))
+        MockPicard.create(os.path.join(self.bin,"picard"))
+        MockGtf2bed.create(os.path.join(self.bin,"gtf2bed"))
+        MockRSeQC.create(os.path.join(self.bin,"infer_experiment.py"))
+        MockRSeQC.create(os.path.join(self.bin,"geneBody_coverage.py"))
+        MockQualimap.create(os.path.join(self.bin,"qualimap"))
+        MockMultiQC.create(os.path.join(self.bin,"multiqc"))
+        os.environ['PATH'] = "%s:%s" % (self.bin,
+                                        os.environ['PATH'])
+        # Make mock analysis project
+        p = MockAnalysisProject("PJB",("PJB1_S1_R1_001.fastq.gz",
+                                       "PJB1_S1_R2_001.fastq.gz",
+                                       "PJB2_S2_R1_001.fastq.gz",
+                                       "PJB2_S2_R2_001.fastq.gz"),
+                                metadata={ 'Organism': 'Chinese hamster' })
+        p.create(top_dir=self.wd)
+        # Set up and run the QC
+        runqc = QCPipeline()
+        runqc.add_project(AnalysisProject(os.path.join(self.wd,"PJB")),
+                          fetch_protocol_definition("standard"),
+                          multiqc=True)
+        status = runqc.run(fastq_screens=self.fastq_screens,
+                           star_indexes=
+                           { 'chinese_hamster': '/data/chok1/star_index' },
+                           annotation_bed_files=
+                           { 'chinese_hamster': self.ref_data['chok1']['bed'] },
+                           annotation_gtf_files=
+                           { 'chinese_hamster': self.ref_data['chok1']['gtf'] },
+                           poll_interval=POLL_INTERVAL,
+                           max_jobs=1,
+                           runners={ 'default': SimpleJobRunner(), })
+        self.assertEqual(status,0)
+        # Check QC metadata
+        qc_info = AnalysisProjectQCDirInfo(
+            os.path.join(self.wd,"PJB","qc","qc.info"))
+        self.assertEqual(qc_info.protocol,"standard")
+        self.assertEqual(qc_info.protocol_specification,
+                         "standard:'Standard QC for single and paired-end data':"
+                         "seq_reads=[r1,r2]:index_reads=[]:qc_modules=[fastq_screen,fastqc,picard_insert_size_metrics,"
+                         "qualimap_rnaseq,rseqc_genebody_coverage,rseqc_infer_experiment,sequence_lengths]")
+        self.assertEqual(qc_info.organism,"Chinese hamster")
+        self.assertEqual(qc_info.seq_data_samples,"PJB1,PJB2")
+        self.assertEqual(qc_info.fastq_dir,
+                         os.path.join(self.wd,"PJB","fastqs"))
+        self.assertEqual(qc_info.fastqs,
+                         "PJB1_S1_R1_001.fastq.gz,"
+                         "PJB1_S1_R2_001.fastq.gz,"
+                         "PJB2_S2_R1_001.fastq.gz,"
+                         "PJB2_S2_R2_001.fastq.gz")
+        self.assertEqual(qc_info.fastqs_split_by_lane,False)
+        self.assertEqual(qc_info.fastq_screens,
+                         "model_organisms,other_organisms,rRNA")
+        self.assertEqual(qc_info.star_index,"/data/chok1/star_index")
+        self.assertEqual(qc_info.annotation_bed,self.ref_data['chok1']['bed'])
+        self.assertEqual(qc_info.annotation_gtf,self.ref_data['chok1']['gtf'])
+        self.assertEqual(qc_info.cellranger_version,None)
+        self.assertEqual(qc_info.cellranger_refdata,None)
+        self.assertEqual(qc_info.cellranger_probeset,None)
+        # Check output and reports
+        for f in ("qc",
+                  "qc_report.html",
+                  "qc_report.PJB.zip",
+                  "multiqc_report.html"):
+            self.assertTrue(os.path.exists(os.path.join(self.wd,
+                                                        "PJB",f)),
+                            "Missing %s" % f)
+        # Check collated Picard insert sizes
+        collated_insert_sizes = os.path.join(self.wd,
+                                             "PJB",
+                                             "qc",
+                                             "insert_sizes.chinese_hamster.tsv")
+        self.assertTrue(os.path.exists(collated_insert_sizes),
+                        "Missing collated insert sizes TSV")
+        with open(collated_insert_sizes,'rt') as fp:
+            self.assertEqual(fp.read(),
+                             """#Bam file	Mean insert size	Standard deviation	Median insert size	Median absolute deviation
+PJB1_S1_001.bam	153.754829	69.675347	139	37
+PJB2_S2_001.bam	153.754829	69.675347	139	37
+""")