Closed statOmics closed 4 years ago
I assume msqrob.R is the main function?
I'll fork locally and to see if it works for me. I'll then send a pull request later making some edits.
@ococrook Yes. In the vignette folder you can find a markdown with an example
Great, thanks!
I get stuck at line 150, which package is combineFeatures exported from?
combineFeatures
come from MSnbase
, but is to be replaced by Features::aggregateFeatures
.
I will be evaluating the model class structure asap. I think there is some scope for simplification, but haven't been able to check in details so far.
I feel so to. I implemented it as such based on the discussion at the developers day to have a StatModel object for every protein. I however would prefer the StatModel Object to be an object with the models of all proteins. It could be a list or a DataFrame with a slot params for all models or parameters, a slot for the type of each model, a slot for all posterior variances and a slot for the posterior degrees of freedom. I guess that we would be able to have a row in the rowData the is a DataFrame itself, right?
Laurent,
Does that mean that I have an old Features package because mine does not have the aggregateFeatures function...
Lieven
@statOmics I think you must have an old version. I re-installed fresh and it works for me.
check that everyone works locally for you by devtools::install() in the package directory. Then require(msqrob2).
Let me know if you can't compile the .rmd file
There are bunch of Errors/Warnings to fix - I should get around to them over the weekend
I had already updated it in the mean time and everything works...
@lgatto @ococrook @ndmeulem First draft of msqrob2 package is functional and can be installed via BiocManager::install("statomics/msqrob2")
Works on simple 3 by 3 cptac.Rmd example.
Please comment on structure.
@ndmeulem Please evaluate on more complex examples and compare with respect to how many proteins that we loose.