"GOplot" and problem with making an integrated matrix

Question

[farzaneafzali]

Hi everybody!

Recently I became familiar with a new Gene Ontology plotting library in R called "GOplot". There is a tutorial in the developer site but from the very first point i have some problems with it. I put the tutorial in this question so you can review it: http://wencke.github.io/

1) There is no function in R like:

 install.github('wencke/wencke.github.io')

2) If you skip this part you go through making dataframes. I don't know how to make mentioned dataframes in toy example part of tutorial for my microarray analysis. I would be really appreciate if you can help me with the second problem specially.

Thanks a lot.

[hel226]

Hi farzaneafzali I am facing the same problem. Did you figure out how to solve it. Thanks, Hossam

[zx8754]

What exactly is the problem, please clarify.

[hel226]

Thank you zx8754 for kind concern. I am beginner with R and i can load my data as .CSV file but the developers are using .rda format which i cannot created i have hard time to do this, can you help me?

[zx8754]

They are just R data file formats, once you read them in they become R objects. Then you can compare with your CSV files. Please see for more info:

• R data formats: RData, Rda, Rds etc

[hel226]

Thank you so much it worked!

[farzaneafzali]

Hi and sorry for the late reply. Recently, I have updated my R version into 3.5 and then updated all my packages. Then, the installation process took place. But, I still have problem with making

circ <- circle_dat(EC$david, EC$genelist)

out of my data. It gives me this error:

Error in `$<-.data.frame`(`*tmp*`, "genes", value = character(0)) : 
  replacement has 0 rows, data has 2203

I will post my problem as the new question now.

Answer 1

[zx8754]

1. You have a typo, try: devtools::install_github("wencke/wencke.github.io"), note that you need to have devtools package installed.

2. Data is within the package, you get them as noted in the tutorial data(EC), it is a list of dataframes.

The toy example
GOplot comes with a manually compiled data set....

[yatta.b]

Hello, I am trying to run GOplot with R. I was able to import my data and make the initial figures. However, when I run the last command (pasted below): I get an error message. Could anyone help me to solve this problem?

circ <- circle_dat(david, genelist) genes <- genes process <- process chord <- chord_dat(circ, genes, process) chord <- chord_dat(data = circ, genes = genes, process = process) GOChord(chord, space = 0.02, gene.order = 'logFC', gene.space = 0.25, gene.size = 5)

I get this error message. Error in $<-.data.frame(*tmp*, "x.start", value = c(0, 0.133335565891055, : replacement has 52 rows, data has 46

Please, zx8754 could you help me with this problem?

[zx8754]

This is a popular error message, see:

• Error - replacement has [x] rows, data has [y]

If this doesn't help you to solve your issue, post a new question (not a comment).

[yatta.b]

Thank you very much!

[jaycal95]

Hi yatta.b - did you ever figure out the reason for this? I am having the same issue and cannot understand why, even with reference forums on the general issue

[georgia.katsoula]

Hi jaycal95,

I had the same problem and I think here it is caused by processes not related to any of your selected genes. Adding this code snippet before the GOchord solved it for me.

chord <- chord_dat(data = circ, genes = genes, process = process)
which(rowMeans(chord) == 0)
which(colMeans(chord) == 0)

    #Eliminate genes related to no processes
    if(length(which(rowMeans(chord) == 0) != 0)){
        paste("one or more genes are not related any of the processes")
        chord <- chord[-which(rowMeans(chord) == 0),]
        }else {
        paste("all genes are related to at least one process")
    }

    #Eliminate processes genes that contain no genes
    if(length(which(colMeans(chord) == 0) != 0)){
        paste("one or more processes are not related any of the selected genes")
        chord <- chord[,-which(colMeans(chord) == 0)]
        }else {
        paste("all processes have at least one related gene")
    }

[farzaneafzali]

Thanks Z I know where TOY DATA is :D I just want to make one of them for myself. enfact in know the code for making dataframe out of my data but it gives me an error when i want to proceed with this code

circ <- circle_dat(go, tn)

This is the code i use for entering the excel files into R and make them datafarme:

go <- data.frame(read.xlsx(file.choose(), colNames=TRUE))

[zx8754]

Then load the toy data, and make your own, and compare the output of str().