CAFE5 rapidly evolving genes

[Jonathan Velotta]

Are only 'significant' gene families designated as 'rapidly evolving' in the CAFE5 gamma modeling, and if so, why would one want to assess non-significant gene family expansion/contraction?

[Hahn, Matthew]

Hi,

“Rapidly evolving” families are those that are significantly faster than the mean rate (on one branch or multiple), while there can still be substantial changes in lots of other important families. So significance is not (just) about the number of changes, but the number relative to all the other families, and you might still want to know about families that have gained and lost genes on a particular lineage.

Matt

On Mar 20, 2024, at 11:13 AM, Jonathan Velotta <jonathan...@gmail.com> wrote:

Are only 'significant' gene families designated as 'rapidly evolving' in the CAFE5 gamma modeling, and if so, why would one want to assess non-significant gene family expansion/contraction?

--
You received this message because you are subscribed to the Google Groups "hahnlab-cafe" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hahnlabcafe...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hahnlabcafe/9ebe63a0-8df0-4c4a-bc80-a152e187112bn%40googlegroups.com.

[Jonathan Velotta]

Hi Matt,

Thanks for your quick reply. In terms of output then, am I correct that significance (p< 0.05) is assigned to rate of evolution, while expansion/contraction alone is not assigned a p-value? I am trying to determine in which output I can find rapidly evolving gene families vs. those with simple expansion/contraction.

Thanks again,

Jon

--

Jonathan Velotta

Assistant Professor

Department of Biological Sciences

University of Denver

2101 E Wesley Ave

Denver, CO 80210

Office: SGM 280

Twitter: @JonVelotta

jvelotta.com

he/him

[Hahn, Matthew]

Yes, exactly right: there is no way to assign a p-value to expansion or contraction alone. That being said, of course the rate is inferred from the numbers of expansions and contractions!

Matt

[Jonathan Velotta]

Wonderful. And thanks for creating the tool!