created by Geraldine_VdAuwera
on 2014-12-06
Warning: the following content may shock or distress our more sensitive users, as we discuss the cold-blooded elimination of some tools from the GATK.
Alright, now that I’ve got your attention (hopefully — if not, what does it take?), here’s the deal. We have got to a point where the GATK is a widely, even massively used toolkit (thanks to you, dear users). And it’s pretty darn robust — it’s what the Broad’s Genomic Platform uses in production to churn out exomes like there’s no tomorrow. But it has technical limitations that are 1) a frequent source of pain on your end and 2) increasingly hampering development of new methods on our end.
The good news is that we have a plan for addressing (read: blasting away) these limitations. But part of this plan will involve streamlining GATK by getting rid of tools that are not useful or are inferior to alternative tools from other packages that we’re not trying to compete with (e.g. Picard tools).
Some tools that are safe from elimination: all the tools used in the Best Practices, and a couple of utilities that we use a lot ourselves. But everything else is up for review — and that’s where you come in: we need your input to decide what to keep, what to throw away, and what to consider rewriting from scratch (yep, this is an option).
[This link](https://www.surveymonkey.com/s/RN8DR3D) will take you to a SurveyMonkey page that lists the tools currently on the chopping block:
[https://www.surveymonkey.com/s/RN8DR3D](https://www.surveymonkey.com/s/RN8DR3D)
Act now to save your favorite non-BP tools! Or help us get rid of the crud. Whichever way you want to look at it, we appreciate your feedback!
Updated on 2014-12-06
From TechnicalVault on 2014-12-08
The survey feels like it’s missing an important column aka “never heard of it”. Otherwise the non-users will either inflate the keep or remove columns depending on their philosophy.
From Geraldine_VdAuwera on 2014-12-08
The intention was to have people just not check any box in that case, so they would only have to put effort into addressing the tools they do care about. I’ll try to clarify the instructions to that effect.
From mard on 2015-02-10
Hi Geraldine, would you know when the results of the survey will be posted? As am particularly keen to hear what’s going to happen to DepthOfCoverage as there’s no other tool that I know of that can count fragments.
From Geraldine_VdAuwera on 2015-02-10
Ah, I kind of forgot that people wanted to see the results, sorry. Not all that many people responded (about ~60) which to me confirms that people don’t care so much about the non-core tools. That said, the people who did respond clearly cared very much about the sequence QC tools such as DepthOfCoverage and DiagnoseTargets. We are not sure yet whether those will stay as they are or will be rewritten (there are some usage issues with both), but rest assured that their functionality will be retained in one form or the other. The plan is, as far as those tools are concerned, some things may be gained, but nothing will be lost.