RELAX is having a "standard issue"

[omario7777]

I am trying to run RELAX but keep getting this issue. I would greatly appreciate a fix

aocc/1.3.0(13):ERROR:150: Module 'aocc/1.3.0' conflicts with the currently loaded module(s) 'aocc/2.1.0'
aocc/1.3.0(13):ERROR:102: Tcl command execution failed: conflict aocc

openmpi/gnu/3.1.6(17):ERROR:150: Module 'openmpi/gnu/3.1.6' conflicts with the currently loaded module(s) 'openmpi/gnu/4.1.0'
openmpi/gnu/3.1.6(17):ERROR:102: Tcl command execution failed: conflict openmpi

Check errors.log for execution error details.

[spond]

Dear @omario7777,

Where are you seeing this? Datamonkey or your own system?

Best,
Sergei

[omario7777]

Hello,

I am using data monkey. I have run into this issue on my mac and two other windows pcs. I don't believe it is an issue on my end.

[spond]

Dear @omario7777,

Would you happen to have the Datamonkey job URL? @stevenweaver : can you take a look at a possible internal misconfiguration?

Best,
Sergei

[omario7777]

Hello,

So sorry for the late response. Here is the link: https://www.datamonkey.org/relax/67ce93bbc7c3b6a8ab8185b6

[stevenweaver]

Dear @omario7777,

Since you’re running RELAX on the website, follow these steps:
1. Click on the branches in the tree visualization.
2. Assign at least one branch as “Test” and at least one as “Reference.”

In this specific job, all branches were set to TEST.

Additionally, to confirm the exact error, click the “Download Log” button to view the full error details. The site was written to expect errors written to stderr/stdout, but HyPhy captures hyphy specific errors and writes them to the same log as the rest of the analysis, hence the confusion there. We hope to address this in a future release of Datamonkey.

Best,
Steven

[omario7777]

Hello Steven,

I have been selecting the whole tree. It runs when I select a branch as "reference" and another branch as "test" as you stated. What if I wanted to do that for the whole tree? Is it possible to reference the tree and test the whole tree at the same time? (I'm very very new to RELAX).

Thanks!

[stevenweaver]

Dear @omario7777,

RELAX requires a distinction between "test" and "reference" branches to compare selective pressures. If you assign the entire tree as both the test and reference set, the analysis loses this contrast, making it uninformative.

However, if you are interested in exploring selection heterogeneity across all branches without predefined reference and test sets, you could use the General Descriptive model in RELAX. This model assigns a separate relaxation/intensification parameter 𝑘 to each branch, allowing you to assess variation in selective pressures throughout the tree without considering a single foreground group.

Best,
Steven