search

biocore / biom-format

The Biological Observation Matrix (BIOM) Format Project
http://biom-format.org
Other
90 stars 95 forks source link

method to convert biom counts to ranked abundances #645

Closed gregcaporaso closed 8 years ago

gregcaporaso commented 8 years ago

I'm working on an analysis where I'd like to identify the most abundant through least abundant observations on a per-sample basis. Is this something that others would like to see in BIOM? For example:

screenshot 2015-10-14 06 34 19

cc @wasade

If so, should it be it's own method, or would it fit better in an existing method? (Or better yet, is it currently possible and I just don't know about it?)

wasade commented 8 years ago

:+1: On Oct 14, 2015 07:35, "Greg Caporaso" notifications@github.com wrote:

I'm working on an analysis where I'd like to identify the most abundant through least abundant observations on a per-sample basis. Is this something that others would like to see in BIOM? For example:

[image: screenshot 2015-10-14 06 34 19] https://cloud.githubusercontent.com/assets/192372/10485187/b08fd396-723d-11e5-897a-44e12a0ab3dc.png

cc @wasade https://github.com/wasade

— Reply to this email directly or view it on GitHub https://github.com/biocore/biom-format/issues/645.

antgonza commented 8 years ago

I like it but what about sorting vs. changing the values, I could see some user getting confused and using the resulting biom for other analysis?

gregcaporaso commented 8 years ago

I need the values changed for the analysis that I want to do (and there is no way to achieve a similar result with sorting anyway as the ranking is on a per-sample basis). We can't really protect users against mis-using their files.

@wasade, what do you think - part of an existing method, or a new method?

antgonza commented 8 years ago

OK, thanks for the explanation.

wasade commented 8 years ago

I believe this can be expressed as a transform (but may be wrong about that). I think it's generality suggests it should be an exposed api method that makes a call to the transform method. Does that make sense? On Oct 14, 2015 9:07 AM, "Greg Caporaso" notifications@github.com wrote:

I need the values changed for the analysis that I want to do (and there is no way to achieve a similar result with sorting anyway as the ranking is on a per-sample basis). We can't really protect users against mis-using their files.

@wasade https://github.com/wasade, what do you think - part of an existing method, or a new method?

— Reply to this email directly or view it on GitHub https://github.com/biocore/biom-format/issues/645#issuecomment-148078845 .

gregcaporaso commented 8 years ago

Yep, that makes sense. Thanks!

On Wed, Oct 14, 2015 at 8:18 AM, Daniel McDonald notifications@github.com wrote:

I believe this can be expressed as a transform (but may be wrong about that). I think it's generality suggests it should be an exposed api method that makes a call to the transform method. Does that make sense? On Oct 14, 2015 9:07 AM, "Greg Caporaso" notifications@github.com wrote:

I need the values changed for the analysis that I want to do (and there is no way to achieve a similar result with sorting anyway as the ranking is on a per-sample basis). We can't really protect users against mis-using their files.

@wasade https://github.com/wasade, what do you think - part of an existing method, or a new method?

— Reply to this email directly or view it on GitHub < https://github.com/biocore/biom-format/issues/645#issuecomment-148078845>

.

— Reply to this email directly or view it on GitHub https://github.com/biocore/biom-format/issues/645#issuecomment-148081765 .