bimsbbioinfo / deconvr Goto Github PK

• 1. Add a new description
• 2. Design a logo
• 3. Organize and re-write some sections in the vignette

In deconvModel.R, the parameter space for nu is specified as seq(0.25, 0.5, 0.75) (line). However, this will only return 0.25 as 0.5 is the end of the sequence and 0.75 is the step size. It looks like the intent was that this is the other way round. I will submit a pull request to fix this.

We may need to upload the data as a new format or, create an ExperimentHub data package or, update from external sources. The current size is now over 9MB

• The doParallel package needs to be added to the deconvRManual.Rmd file
  

İf probe_id_locations is a data.frame, this error returns

I came by the package home page to reference the package for my thesis and noticed the page does not show the package author information at the bottom of the page (see screenshot).

You probably just need to rebuild the page. However it seems like the version displayed on the home page differs from both the version specified in the package description, as well as the version on Bioconductor.

The BiocCheck function was used to determine if the deconvR complies with the Bioconductor package guidelines. deconvR failed in the test and, the issues needed to be solved detected

Summary:
ERROR count: 1
WARNING count: 2
NOTE count: 6
For detailed information about these checks, see the BiocCheck vignette,
https://bioconductor.org/packages/3.13/bioc/vignettes/BiocCheck/inst/doc/BiocCheck.html#interpreting-bioccheck-output

• The codes needed to be formatted
• Comments needed to be organized
• BiocCheck warning don't use is() instead of class(), is there any other way to solve this warning?
• BSmeth2Probe name is appropriate or not? If not, need to be changed
• CRAN allows documentation size up to 5 MB, deconvR has documentation size 6.3 MB. Is this an issue, if it is, how it can be solved?

For more information : bioconductor code styling

deconvR passed the BiocCheck but, there are still some issues that needs to be solved

in the details section, we should have a text similar to the one below:

We have this function to check if deconvolution brings anything useful on top of the basic bimodal methylation profiles. in case of methylation, the reference matrix usually follows a bimodal distribution an taking the average of the rows of methylation matrix might give a pretty similar profile to the bulk methylation profile you are trying to deconvolute. If the deconvolution is useful partial R-squared is high.

• R Markdown need to be organized according to BiocStyle Vignette
• Lines should be simpler and shorter

For more information BiocStyle Vignette

please add irem to the authors. If you are going to submit this to bioconductor please add one more maintainer. If that's not the aim, leave it as is.

• 1. unable to find an inherited method for function ‘findOverlaps’ for signature ‘"GRanges", "data.frame"’
  (This error is returned if probe_id_locations is a data.frame)
• 2. print(paste(...)) argumants must be replaced with appropriate function
• 3. simulateCellMix returns 0 bug need to be fixed
• 4. deconvolute nnls executing %dopar% sequentially: no parallel backend registered
• 5. since the dependencies took so long to install, find a way to reduce this time
• 6. Supported methylkit objects needed to be checked
• 7. Get rid of the for loops, they took so much time