David Steinberg

Thanks! This thread was started specifically to address how to model the TCGA data! Is there a minimum change to the biosample message that would allow this? Perhaps we might...

It works for me, I think if we go with a tag-bag approach a logical OR would satisfy most use cases.

@mbaudis with the characteristics we have improved the granularity of describing phenotypes, however, following on our conversations Monday, I believe we need to add a `tissue_site` that allows one to...

Thanks @jpdna ! Two problems have come together here. First, [RTD is failing](https://github.com/ga4gh/ga4gh-server/issues/1599) so our "latest" docs are actually aging quickly. The `species` field was added recently and so the...

If you're building from source the `latest` RTD has been updated now! http://ga4gh-reference-implementation.readthedocs.io/en/latest/datarepo.html#command-line

This is a bug when converting from SAM to BAM using pysam, which is done in `prepare_compliance_data`. ``` import pysam input = pysam.AlignmentFile("../compliance/test-data/brca1_HG00101.sam") output = pysam.AlignmentFile("out.bam", 'wb', header=input.header) print input.header.get('PG')[1].get('PP')...

This has been pushed up to pysam and they say they are working on a fix.

We should try to find a way to parallelize some of the tests. Not all of them are written to be reentrant, so we might just run each test directory...

Yeah I've tried it out once or twice. Hence why I think we could just split them across processes in the shell, though Travis instances are pretty light on resources....

Related https://github.com/ga4gh/ga4gh-server/issues/987