Ensure that sample_id and experiment_id are passed when extracting batch info

Author: arteymix

rnaseq_pipeline/sources/geo.py

@@ -103,9 +103,9 @@ def run(self):
             raise RuntimeError('{} GEO record is not linked to SRA.'.format(self.gsm))
         platform, srx_url = samples_info[self.gsm]
         srx = parse_qs(urlparse(srx_url).query)['term'][0]
-        yield DownloadSraExperiment(srx, metadata=self.metadata)
+        yield DownloadSraExperiment(srx, metadata=dict(**self.metadata, sample_id=self.sample_id))
 
-class DownloadGeoSeriesMetadata(RerunnableTaskMixin, luigi.Task):
+class DownloadGeoSeriesMetadata(TaskWithMetadataMixin, RerunnableTaskMixin, luigi.Task):
     """
     Download a GEO Series metadata containg information about related GEO
     Samples.
@@ -138,7 +138,7 @@ def run(self):
         gsms = collect_geo_samples(self.input().path)
         if not gsms:
             raise ValueError('{} has no related GEO samples with RNA-Seq data.'.format(self.gse))
-        yield [DownloadGeoSample(gsm) for gsm in gsms]
+        yield [DownloadGeoSample(gsm, metadata=self.metadata) for gsm in gsms]
 
 @requires(DownloadGeoSeriesMetadata, DownloadGeoSeries)
 class ExtractGeoSeriesBatchInfo(luigi.Task):

rnaseq_pipeline/sources/sra.py

@@ -157,7 +157,7 @@ def run(self):
         # on the number of mates per spot
         is_paired = (self.sample_id in sra_cfg.paired_read_experiments) or (run.spots_with_mates > 0) or (run.LibraryLayout == 'PAIRED')
 
-        yield DumpSraRun(run.Run, self.srx, paired_reads=is_paired, metadata=self.metadata)
+        yield DumpSraRun(run.Run, self.srx, paired_reads=is_paired, metadata=dict(**self.metadata, sample_id=self.sample_id))
 
 class DownloadSraProjectRunInfo(TaskWithMetadataMixin, RerunnableTaskMixin, luigi.Task):
     """

rnaseq_pipeline/tasks.py

@@ -376,9 +376,9 @@ def requires(self):
         # TODO: Have a generic strategy for extracting batch info that would
         # work for all sources
         if self.external_database == 'GEO':
-            return ExtractGeoSeriesBatchInfo(self.accession)
+            return ExtractGeoSeriesBatchInfo(self.accession, metadata=dict(experiment_id=self.experiment_id))
         elif self.external_database == 'SRA':
-            return ExtractSraProjectBatchInfo(self.accession)
+            return ExtractSraProjectBatchInfo(self.accession, metadata=dict(experiment_id=self.experiment_id))
         else:
             raise NotImplementedError('Extracting batch information from {} is not supported.'.format(self.external_database))