New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
differentialTest
fails with boot = TRUE
#128
Comments
Hi @amorris28 , Thanks for posting this issue! I am having a bit of trouble re-producing this issue. Can you please try a few things for me and let me know what the output is? That will help me diagnose the problem here.
To set custom init values, you just need to provide a vector of values. For example, you could just do Keep me updated! I'll work through this I just need a bit more information to figure it out. |
Okay, so I ran the example for I tried running my dataset again with I tried again with
|
Hi @amorris28 , Is your data available to be shared? If so, would you be willing to send it to me? I'm having trouble diagnosing this and it would be helpful to debug looking at the same data you are encountering the problem with. Bryan |
Hi @amorris28 and @bryandmartin - I've been following this conversation (and similar issues posted by others), as I'm also running into similar problems. Are there any updates? So far I can't even get things to work when setting B=5 and trying to use custom init values. Regardless of the value I use for p in 'init=rep(1,p)' I receive the error: "Error in differentialTest(formula = ~source4, phi.formula = ~1, formula_null = ~1, : Thanks in advance for any insight, |
Hey, @bryandmartin. Just wanted to check on this issue. I emailed you my phyloseq object on 2/6, but I don't know if you receive it. Let me know if you didn't. Planning on submitting this manuscript soon with A |
Hi @bryandmartin -- let me know if you would like help with Hope you're well! Cheers, Amy |
Hi da_analysis_donors_f_b<- differentialTest(formula= ~Prog,
OPTION 2: boot=TRUE, B=5, --> 0 significant taxa da_analysis_donors_f_b<- differentialTest(formula= ~Prog,
OPTION 3: BOOT=TRUE, FAILS Error in if (is.nan(val) || any(phi <= sqrt(.Machine$double.eps)) || any(phi >= : -- devtools::session_info() ─ Packages ────────────────────────────────────────────────────────── [1] /Library/Frameworks/R.framework/Versions/4.2-arm64/Resources/library ───────────────────────────────────────────────────────────────────── |
Hi @amorris28 -- thanks for your patience! Bryan has moved on to a new position and I'm helping to maintain the package going forward. Are you able to send me the data that you sent Bryan? We'll do our best to give you an answer soon. You can find my email in the header at http://statisticaldiversitylab.com/. I'm very sorry for the delay, and I hope your manuscript submission has gone well. Amy |
Thanks so much for sharing your data with me via email, @amorris28. The default I could reproduce an error, but can diagnose it as a user-input error. Here's your original code:
and here is code that runs:
(Notice the capitalization of "LRT".) This issue recently came to our attention in #148 and so actually the amazing @svteichman actually wrote a helpful pull request #151 which now gives more information ("Alternatively, double-check your values for the arguments I have further refined the error message in #155 -- actually I did this independently of your issue as I'm extending the testing options 😸 -- "If no errors were thrown by I believe this resolves the original issue so I'm going to close -- please feel free to reopen if your concern is not fully addressed. To others on this thread also finding this issue with
Thanks all for your patience and feedback -- it helps us grow better 🌽 ! |
I have a dataset that is 24 samples, 12 per treatment. I have aggregated my ASVs at the family level and there are approximately ~300 families. I have tried setting the minimum prevalence to 10% and 20%. I'm testing differential abundance between the two treatments with the variable
treat
which is a binary variable (p
/n
).Here is the command I'm running:
When I run this, it appears to print each model to the Console as it runs using
print.bbdml
. Most of them run fine, except for the occasional model that says this:After it finishes I get this error message:
I have tried running a number of taxa individually using
bbdml
as well and each one I've tried has worked:Based on issue #121 and #123 , I tried reinstalling the
trust
andoptimr
packages like this:Still getting the same error message.
I have not tried setting custom
init
values, but I don't know how to do that.Any ideas how to get the parametric bootstrap to work?
Thanks!
sessioninfo()
:The text was updated successfully, but these errors were encountered: