daylily
daylily copied to clipboard
Daylily-Informatics
A NGS analysis framework for WGS data, which automates the entire process of spinning up AWS EC2 spot instances and processing FASTQ to snvVCF in <60m, for dollars a sample and achieving Fscores of 0....
So that when this is run, it is not blindly pulling head. Consider a reclone script from the core ~/projects/ repo created on the headnode, whicih could wrap the git...
ie: add to the BAM metadata, lib prep, institution, etc. Pull from sample info in analysis_manifest.csv (and better, from the sample and units files in v8)
The current meta report files all are named in such a way they will clash if combined from multiple pipelines. This is bad news bears (ie: benchmarks.tsv)
Daylily jobs used to run on multiple different vcpu configurations, but then I simplified to i192 when it struck me the per vcpu cost of smaller instances was often much...
The most common tools to be used willbe deepvariant and bwa2a, might as well save the folks pilling this every time a cluster is built. ALSO- add clear section in...
- make version naming consisten with s/w repo naming, be direct copies, and add version to bucket naming.
- remove the `resources/lib` and `/fsx/data/lib` directories from the project and test.
Metadata
Owner
Metadata
A NGS analysis framework for WGS data, which automates the entire process of spinning up AWS EC2 spot instances and processing FASTQ to snvVCF in <60m, for dollars a sample and achieving Fscores of 0....