(inactive) BASiNET

[EricIto]

• Repository: https://github.com/EricIto/BASiNET

Confirm the following by editing each check box to '[x]'

• [x] I understand that by submitting my package to Bioconductor, the package source and all review commentary are visible to the general public.

• [x] I have read the Bioconductor Package Submission instructions. My package is consistent with the Bioconductor Package Guidelines.

• [x] I understand that a minimum requirement for package acceptance is to pass R CMD check and R CMD BiocCheck with no ERROR or WARNINGS. Passing these checks does not result in automatic acceptance. The package will then undergo a formal review and recommendations for acceptance regarding other Bioconductor standards will be addressed.

• [x] My package addresses statistical or bioinformatic issues related to the analysis and comprehension of high throughput genomic data.

• [x] I am committed to the long-term maintenance of my package. This includes monitoring the support site for issues that users may have, subscribing to the bioc-devel mailing list to stay aware of developments in the Bioconductor community, responding promptly to requests for updates from the Core team in response to changes in R or underlying software.

I am familiar with the essential aspects of Bioconductor software management, including:

• [x] The 'devel' branch for new packages and features.
• [x] The stable 'release' branch, made available every six months, for bug fixes.
• [x] Bioconductor version control using Git (optionally via GitHub).

For help with submitting your package, please subscribe and post questions to the bioc-devel mailing list.

[bioc-issue-bot]

Hi @EricIto

Thanks for submitting your package. We are taking a quick look at it and you will hear back from us soon.

The DESCRIPTION file for this package is:

Package: BASiNET
Title: Classification of mRNA and lncRNA sequences using complex network theory
Version: 0.99.0
Author: Eric Augusto Ito
Maintainer: Eric Augusto Ito <ericaugustoito@hotmail.com>
Description: It makes the creation of networks from sequences of RNA, with this is done the abstraction of characteristics of these networks with a methodology of threshold for the purpose of making a classification between the classes of the sequences.
Depends: R (>= 3.4.1)
License: GPL-3
Encoding: UTF-8
LazyData: true
Imports: igraph, seqinr, RWeka, rmcfs, randomForest, rgl
biocViews: Software, BiologicalQuestion, GenePrediction
RoxygenNote: 6.0.1
Suggests: knitr,
    rmarkdown
VignetteBuilder: knitr

[mtmorgan]

Use Biostrings rather than seqinr. Your vignette must be a full exposition of package use, with fully evaluated R code chunks. Make these changes and submit a brief note indicating what you have done. Your package will then be considered for review.

[EricIto]

I've changed the package as you advised, now I'm using Biostrings with the readBStringSet () function. And I also changed the vignette, now I have detailed more like that using function and parameters

[bioc-issue-bot]

Your package has been approved for building. Your package is now submitted to our queue.

IMPORTANT: Please read the instructions for setting up a push hook on your repository, or further changes to your repository will NOT trigger a new build.

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the following build report for more details:

http://bioconductor.org/spb_reports/BASiNET_buildreport_20171128080738.html

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

6642c74 bump version

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the following build report for more details:

http://bioconductor.org/spb_reports/BASiNET_buildreport_20171130112301.html

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

eedd2f9 Reenviu

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the following build report for more details:

http://bioconductor.org/spb_reports/BASiNET_buildreport_20171205151337.html

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

1fe9954 bump version

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "skipped, ERROR, WARNINGS". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the following build report for more details:

http://bioconductor.org/spb_reports/BASiNET_buildreport_20171205153822.html

[EricIto]

Dear Bioconductor, Dear @mtmorgan,

I have three warning/error :

• checking whether package BASiNET can be installed ... [7s/5s] WARNING Found the following significant warnings: Warning: 'rgl_init' failed, running with rgl.useNULL = TRUE

• Warning in rgl.init(initValue, onlyNULL) : RGL: unable to open X11 display Warning: 'rgl_init' failed, running with rgl.useNULL = TRUE

• ERROR: dependencies RWeka, rmcfs are not available for package BASiNET

  Any suggestions on what I can do ? I tried to fix these errors but the problem is that on my computer these errors do not happen

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

7088570 version

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "WARNINGS, skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the following build report for more details:

http://bioconductor.org/spb_reports/BASiNET_buildreport_20171207133120.html

[dvantwisk]

Both problems with the error and warnings are on our end. More specifically, CRAN binaries are not available for R 3.5 on the Mac yet, so this error cannot be corrected until the binaries are available. You can ignore the merida2 build error for now. You can also ignore the warning on malbec2 involving rgl. This is a warning we are still working in resolve on our end. We will start reviewing your package now.

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

2440228 Bump version and fix graph3D a1ad566 Merge branch 'master' of https://github.com/EricIt...

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "WARNINGS, skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the following build report for more details:

http://bioconductor.org/spb_reports/BASiNET_buildreport_20171217131839.html

[dvantwisk]

I've been reviewing your package and I'd just like you to correct this issue before I continue.

vignettes

• [REQUIRED] As it stands now, the only runnable code in your vignette are some knittr commands. Please put your variables and classfication() call into a code block so that this sections can be run.

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

bc04ac1 Correction of notes, change of vignette The examp...

[EricIto]

I put the examples in a block of code. See if I did it correctly please

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the build report for more details.

[dvantwisk]

You need to write a code block as follows: ``{r} a <- 1:10 sum(a) \``` You can run the commandR CMD Stangle ` to create an R file that shows all runnable code in the vignette. You can use this to check whether you got it right.

Before I give you the rest of the review, I do need to confirm that you know that BASiNET is already a package on CRAN. Once your package is accepted here, you must remove the package from CRAN since having the same packages on two concurrent repositories can cause problems. Do you accept this?

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

cfb0043 Correction of the code block

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the build report for more details.

[EricIto]

I agree, when my package is accepted in BASiNET I will remove from CRAN

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

d47e641 Fixed some bugs

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the build report for more details.

[dvantwisk]

Here is a review to get you started. There may be more things I need to bring up after you make these changes. I'm having trouble with all the matrices you are creating in classification.R:65-136. This comes off to me as an odd way to organize this data, but as it stands, most of the functions that help classification() that you have created are not vectorized and changing this method of organization may be a rather large overhaul. I'm still deciding what to do for this part. In the meantime, please make the following changes:

GENERAL:

• [REQUIRED] Add .DS_Store to .gitignore
• [REQUIRED] In your Roxygen headers be sure to be specific about what type of argument a parameter required (e.g. numeric(), logical(1), data.frame). Also specify specific exceptable files types, etc.
• [CONSIDER] return() at end of functions generally not needed.

DESCRIPTION:

• [REQUIRED] update required R version to 3.5.0

R/classification.R

• [REQUIRED] Use match.arg() to improve style of function declaration. For example, the parameter interval.type can have two values "J48" and "RF". The parameter in the function declaration can be written as interval.type=c('J48', 'RF'). The value that the user uses can be obtained with classifier <- match.arg(classifier). The first value (in this case J48) is the default argument. Using this method makes it clear to users what the allowed inputs to a function are and may also make error catching easier.
• [REQUIRED] L35: For parameters with a default value you can use e.g. step = 1 inside the function argument instead of calling missing() to set values.
• [REQUIRED] L76, and many other places: use seq_along() or seq_len() instead of 1:.... This avoids the potentially unexpected case where 1:0 returns 1 0 and in addition is also slightly faster.
• [REQUIRED] Use message() instead of print() for diagnostic messages.
• [REQUIRED] L64-140: This block looks very inefficient. In would be far better to not iteratively add on columns to the matrices and instead find a way to vectorize measures(). It may be better to create seperate matrices for mRNA, lncRNA, and srcRNA instead of iterating across numClass and using the long if-else statments in L78-97.
• [REQUIRED] L200-202: data$CLASS[seq_len(seqMRNA)] <- 'mRNA'

tests

• [CONSIDER] Adding tests to your package using the testthat package.

vignettes

• [REQUIRED] change name of vignette file to something informative as it will have this name when searching for it using vignette().

[bioc-issue-bot]

Received a valid push; starting a build. Commits are:

1e23e7e The changes were made according to the review

[bioc-issue-bot]

Dear Package contributor,

This is the automated single package builder at bioconductor.org.

Your package has been built on Linux, Mac, and Windows.

On one or more platforms, the build results were: "skipped, ERROR". This may mean there is a problem with the package that you need to fix. Or it may mean that there is a problem with the build system itself.

Please see the build report for more details.

[EricIto]

About this: [REQUIRED] L64-140: This block looks very inefficient. Would be better iteratively add to columns to the matrices and instead find a way to vectorize measures (). It may be better to create seperate matrices for mRNA, lncRNA, and srcRNA instead of iterating across numClass and using the long if-else statments in L78-97.

In this part of the code the measurement results are extracted and saved, these measurements are taken for each sequence, for each sequence there is an iteration of applying thresholds and extracting the measurements, this iteration is repeated N times and varies according to the network. That's why I'm adding columns in the matrix when I need it, and this feature that each network has a N of iterations is why I can not change that block of code

[dvantwisk]

At L64-140, I can't give a clear recommendation on how to change this code, so for the time being we can overlook it. Please look into making the other changes though.

[dvantwisk]

Sorry for the wait. I have some more things for you to look into. These are mostly some improvements to some of the issues I had in classification.R. Please implement these changes and message me back when you have. Also feel free to comment if you have any questions, etc.

GENERAL:

• [REQUIRED] add .DS_STORE files in every directory to .gitignore
• [CONSIDER] Removing commented out code.

R/classification.R

• We've been looking at the code in this file and have some recommendations to improve the look and runtime, but without compromising the structure of the package: L74-90: Replace with:

  if (k == 1) {
  message("Analyzing mRNA from number: ")
      seq<-seqMRNA
  } else if (k == 2) {
  message("Analyzing lncRNA from number: ")
  seq<-seqLNCRNA
  } else {            # k == 3
  message("Analyzing sncRNA from number: ")
  seq<-seqSNCRNA
  }

  This is a better way to write these conditionals. L94: Remove limitThreshold <- 0 (this is unneeded) L100-127: Replace with:

          n_new_columns <- limitThreshold - ncol(averageShortestPathLengths)
          m_new <- matrix(nrow = numSeq, ncol = n_new_columns)
          averageShortestPathLengths <- cbind(averageShortestPathLengths, m_new)
          clusteringCoefficient <- cbind(clusteringCoefficient, m_new)
          ...
  
          vector <- sapply(seq_len(limitThreshold), function(t) {
                      net<<-threshold(t, net)
                      measures(net)
          })
  
          cidx <- seq_len(nrow(vector))
          averageShortestPathLengths[aux + x, cidx] <- vector[1,]
          clusteringCoefficient[aux+x, cidx] <- vector[2,]
          ...

  The ... here indicate where to continue writing code in the pattern we've specified. This revised code avoid a for loop which included a cbind within the loop. This operation is very expensive in R. L131-154:

  numCol<-ncol(averageShortestPathLengths)
  
  listMatrix<-list(
      averageShortestPathLengths,clusteringCoefficient,
      standardDeviation,maximum,assortativity,betweenness,degree,
      minimum,motifs3,motifs4)
  listMatrix <- lapply(
      listMatrix, reschedule,
      length(seqMRNA),length(seqLNCRNA),length(seqSNCRNA)
  )
  
  namesMeasure<-c(
      "Average shortest path length", "Cluster Coefficient",
      "Standard deviation", "Maximum", "Assortativity",
      "Betweenness", "Degree", "Minimal", "Motifs 3", "Motifs 4"
  )

  This just cleans up a section of code that repeated itself a lot.

R/threshold.R L28-36: Replace with:

       if(x!=1){
        matriz<-as_adjacency_matrix(net)
        matriz[(matriz==(x-1))]<-0
        net<-graph_from_adjacency_matrix(matriz, mode="undirected")
    }
    return(net)

The first conditional is empty and is, therefore, not needed.

[dvantwisk]

Hello,

Have you been working on the changes that we requested? We plan to have a new Bioconductor release in early May, so If you want your package to be included in that release, it will need to be accepted by mid-April. Please message me back and feel free to ask for any help if it is needed.

[EricIto]

Yes, I've been working. I made the changes you suggested and I made other changes as well. Now I'm doing some tests and then I'm going to upload here

[dvantwisk]

Just messaging you back to check in on your progress. This is another two week notice. Do you want to keep this issue open?

[EricIto]

I have a doubt, in the checklist it speaks like this "The result of R CMD build must be less than 4MB". I made some changes in BASiNET, now it is possible to save a model and load this model so that it does not have to do the training of the data, but only the test of the sequences. I want to make available in the package some models for several species, however with that the size of the package was to 83.1mb. Do you have any suggestions for me?

[dvantwisk]

You'll need to submit the training data as an experiment data package. We don't have any set guidelines for creating one, but I can give you an example of what the package should look like. alpineData https://bioconductor.org/packages/release/data/experiment/html/alpineData.html is an experiment data package that contains large data files to be used as an example for the alpine package https://bioconductor.org/packages/release/bioc/html/alpine.html. You'll need to submit this package similarly to how you submitted this one, but leave a message stating AdditionalPackage: https://github.com/EricIto/BASiNET to show that the package is used in conjunction with another. The closed issue for alpineData shows this https://github.com/Bioconductor/Contributions/issues/69.

[dvantwisk]

Hello,

I'm just sending out this message to those who are still working on their packages. The package needs to be accepted by April 16th to be included in the upcoming Bioconductor 3.7 release.

[dvantwisk]

This issue has been inactive for a while. It's been a while since anything has been pushed to it. If the package is to be included in the 3.7 release, it must be accepted by April 16th, but that includes the time it takes us to review the changes. Please make updates to this package soon or we will need to close this issue for the time being.

[dvantwisk]

I am going to close this issue for the time being since it has been inactive. You may reopen this issue when you are ready to make further changes.

[bioc-issue-bot]

There has been no progress on this issue for an extended period of time and therefore will be closed due to inactivity. You may reopen the issue when you feel you have the time to actively participate in the review/submission process. Please also keep in mind when a package is accepted to Bioconductor a level of active maintenance is expected.

Thank you for interest in Bioconductor.