Re: Planning a Debian Med Covid-19 project

[Steffen Möller <steffen_moeller@gmx.de>]

That is the Virus-side of the disease. Also nice.

On 07.11.20 11:44, Michael Crusoe wrote:
> I would recommend getting as many components of
> http://covid19.genenetwork.org/ <http://covid19.genenetwork.org/>
> packaged for Debian as possible.
>
> Workflows are at
> https://github.com/hpobio-lab/viral-analysis/tree/master/cwl/pangenome-generate
> <https://github.com/hpobio-lab/viral-analysis/tree/master/cwl/pangenome-generate>
>
> Maybe someone can start by identifying the software used and seeing
> which is already packaged for Debian?
>
> Currently the workflows are run on Arvados. How is the packaging of
> Arvados coming along?
>
> -- 
> Michael R. Crusoe
>
> On Fri, Nov 6, 2020, 17:50 Steffen Möller <steffen_moeller@gmx.de
> <mailto:steffen_moeller@gmx.de>> wrote:
>
>     Dear all,
>
>     We have multiple aims. Here are mine:
>
>     a) do something good
>     b) show to the world that our infrastructure is usable
>     c) do something that is not redundant
>     d) be educational
>     e) come up with findings that can be reproduced
>     f) have it all automated, preferably in a way that it can be
>     adapted for
>     other diseases/symptoms/whatever.
>
>     Idea: We do not go for a single paper that we want to reanalyse (would
>     be redundant at best) but get _all_ RNA-seq raw data on Covid-19. I
>     thought about  using PiGx-rnaseq (bcbio is not ready) to get us read
>     counts and the typical gene set enrichment analyses. With that many
>     samples, we should also have options for downstream analyses for which
>     earlier efforts possibly lacked the statistical power.
>
>     How do you feel about this? Should we prepare a Virtual Sprint for it?
>
>     Best,
>
>     Steffen
>