search

biocore / qiime

Official QIIME 1 software repository. QIIME 2 (https://qiime2.org) has succeeded QIIME 1 as of January 2018.
GNU General Public License v2.0
286 stars 266 forks source link

Implement ANCOM, perhaps as part of differential_abundance.py #2090

Open colinbrislawn opened 9 years ago

colinbrislawn commented 9 years ago

This recent paper presents Analysis of Composition of Microbiomes (ANCOM) http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4450248/

Congrats @wdwvt1, @rob-knight! I would love to see this method in qiime!

wasade commented 9 years ago

It's currently in an open PR biocore/scikit-bio#1114; agree it would be great to have available On Oct 13, 2015 9:55 AM, "Colin Brislawn" notifications@github.com wrote:

This recent paper presents Analysis of Composition of Microbiomes (ANCOM) http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4450248/

Congrats @wdwvt1 https://github.com/wdwvt1, @rob-knight https://github.com/rob-knight! I would love to see this method in qiime!

— Reply to this email directly or view it on GitHub https://github.com/biocore/qiime/issues/2090.

colinbrislawn commented 9 years ago

Excellent! The implementation itself does make more sense as part of skikit-bio, and this function can be added as a method in differential_abundance.py.

You can close this issue if you want.

gregcaporaso commented 9 years ago

There is a problem here unfortunately: QIIME 1 won't be updated to use a post 0.2.x version of scikit-bio, so this method won't be available. Unfortunately I think we'll have to wait for QIIME 2 to get moving if we're going to use the scikit-bio implementation (but it should be easy to use the scikit-bio implementation outside of QIIME once that PR is merged).

ebolyen commented 9 years ago

Technically speaking, there is no reason we couldn't backport ANCOM to a 0.2.4 release, but that isn't me volunteering...

gregcaporaso commented 9 years ago

Yep, it all comes down to what we have the bandwidth for. I'm really excited to start using this method, but I think the way to get there is to start focusing on QIIME 2 development.

colinbrislawn commented 9 years ago

I'm comfortable with this being incorporated into QIIME 2. If I need it before then, a colleague or I can run it through scikit-bio.

This is off topic, but what's the timeline for QIIME 2 release? Is your team planning a 1.10.0 release before 2.0.0? Just curious...

gregcaporaso commented 9 years ago

No, we're not planning a 1.10.0 release, but we will do bug fix 1.9.x release if necessary. The plan is that 1.9.x is going to be a long-term support release which we'll continue to support as we're developing, releasing, and supporting early versions in the 2.x.x series of release. We have a call coming up to discuss the plan for the 2.0.0 timeline, and we'll share some information on that with the community after the call (we'll post to the forum and the blog to indicate where that information can be found).

On Tue, Oct 13, 2015 at 3:32 PM, Colin Brislawn notifications@github.com wrote:

I'm comfortable with this being incorporated into QIIME 2. If I need it before then, a colleague or I can run it through scikit-bio.

This is off topic, but what's the timeline for QIIME 2 release? Is your team planning a 1.10.0 release before 2.0.0? Just curious...

— Reply to this email directly or view it on GitHub https://github.com/biocore/qiime/issues/2090#issuecomment-147872818.

gregcaporaso commented 9 years ago

In case folks haven't seen it, the blog post went up here.

gregcaporaso commented 9 years ago

ANCOM is now merged in scikit-bio, so will be included in the upcoming 0.4.1 release.