Jeff Hammerbacher

Results 167 comments of Jeff Hammerbacher

Make example data sets public

Those are a bit large for Google Drive. Nature [recommends](https://www.nature.com/sdata/policies/repositories#imaging) the IDR: cf. their [submission guidelines](https://idr.openmicroscopy.org/about/submission.html). If IDR doesn't want our images, maybe [Dryad](https://datadryad.org)? I dunno. We should ask Anne...

Make example data sets public

The [Cell Image Library ](http://www.cellimagelibrary.org/home) is another alternative

Potential source of validated somatic mutations

Another option: http://txcrb.org/open.html

Test running on Baylor "case 2" dataset

@ryan-williams is this task still in progress?

Add MHCflurry option to MHC predictor

🙏

Measure coverage of BAMs in QC steps

cc @ryan-williams

Make a public bucket or NFS filer available w/ sample data

The sample data should include reference + index and dbSNP/ExAC/COSMIC

Make a public bucket or NFS filer available w/ sample data

Looks like hg38 is available w/ index: https://console.cloud.google.com/storage/browser/genomics-public-data/resources/broad/hg38/v0/

Document command-line args required by pipeline COMMAND

Currently running over @smondet's private DREAM chr20 BAMs: `KETREW_CONFIG=./ketrewdocker/configuration.ml ocaml jh-epidisco.ml pipeline -E jh-epidisco-test --results-path /nfswork/jh-epidisco-test --reference-build=b37 --normal /nfstest1/datasets/training-dream/normal.chr20.bam --tumor /nfstest1/datasets/training-dream/tumor.chr20.bam`

Resolve overlap between epidisco and cohorts

`epidisco` for anything **automatically** generated