Question

MA plot

0

Entering edit mode

Nathalie ▴ 20

@ab7ad8dd

Last seen 19 months ago

France

Hello, I am a beginner in R studio. I try to do a MAplot after a differential analysis. I try this code but I obtain a message of error. I insert a picture of the dataframe. Could you help me ? I find this code : [https://rpkgs.datanovia.com/ggpubr/reference/ggmaplot.html[1] Thank you very much

Code should be placed in three backticks as shown below

dtf <- read.table("Essai.txt")
diffexp <- dtf

library(geneplotter)
library(ggmatplot)
ggmaplot(diffexp, main = expression("Group 1" %->% "Group 2"),
         fdr = 0.05, fc = 2, size = 0.4,
         palette = c("#B31B21", "#1465AC", "darkgray"),
         legend = "top", top = 20,
         font.label = c("bold", 11),
         font.legend = "bold",
         font.main = "bold",
         ggtheme = ggplot2::theme_minimal())


Error in ggmaplot(nana, main = expression("Group 1" %->% "Group 2"), fdr = 0.05,  : 
  could not find function "ggmaplot"

MAplot • 2.7k views

ADD COMMENT • link 22 months ago Nathalie ▴ 20

0

Entering edit mode

Nathalie ▴ 20

@ab7ad8dd

Last seen 19 months ago

France

Thank you for your help. I install the package, I change the names of columns (message error). I run the code and I obtain a new message of error.

Error in data$baseMean + 1 : non-numeric argument to binary operator

ADD COMMENT • link 22 months ago Nathalie ▴ 20

2

Entering edit mode

This likely has to do with the 1st row in your input; these are the names of the columns, thus characters, and not numbers. Fix this by adding the argument header = TRUE to your call when reading in the data. Thus:

dtf <- read.table("Essai.txt", header = TRUE)

Also note that you have comma's as decimal marks, and whereas dots may be expected.... Check the argument dec of read.table (default is dec = ".").

ADD REPLY • link 22 months ago Guido Hooiveld ★ 4.1k

0

Entering edit mode

Thank you for your help. I made the modifications. But I obtain a new error message that I don't understand. Could you help me ?

 dtf <- read.table("Essai.txt", header = TRUE)
str(dtf)``
library(ggpubr) 
library(geneplotter)
library(ggmatplot)
ggmaplot(dtf, main = expression("Group 1" %->% "Group 2"),
         fdr = 0.05, fc = 2, size = 0.4,
         palette = c("#B31B21", "#1465AC", "darkgray"),
         legend = "top", top = 20,
         font.label = c("bold", 11),
         font.legend = "bold",
         font.main = "bold",
         ggtheme = ggplot2::theme_minimal())



  > str(dtf)
'data.frame':   24405 obs. of  3 variables:
 $ baseMean      : num  0.05 -0.05 0.17 0.83 -0.03 0 0.17 0 0.17 -0.71 ...
 $ log2FoldChange: num  -0.88 -0.22 0.21 8.17 2.46 -0.18 3.59 4.14 3.25 -2.38 ...
 $ padj          : num  0.04 0.03 0.38 14.6 0.02 0 0.82 0 0.26 0.57 ...
> library(geneplotter)
> library(ggmatplot)
> library(ggmatplot)
> ggmaplot(dtf, main = expression("Group 1" %->% "Group 2"),
+          fdr = 0.05, fc = 2, size = 0.4,
+          palette = c("#B31B21", "#1465AC", "darkgray"),
+          legend = "top", top = 20,
+          font.label = c("bold", 11),
+          font.legend = "bold",
+          font.main = "bold",
+          ggtheme = ggplot2::theme_minimal())
Error in seq.default(0, max(data$mean), 2) : 
  'to' doit être un nombre fini
In addition: Warning message:
In ggmaplot(dtf, main = expression("Group 1" %->% "Group 2"), fdr = 0.05,  :
  NaNs produced
>

>

ADD REPLY • link 22 months ago Nathalie ▴ 20

3

Entering edit mode

Without having a reproducible example, or access to your data, it is almost impossible to help you. Also note that ggpubr is not a Bioconductor package...

After having had a quick look at your code I noticed some strange things in your input: the values are really low for a baseMean..., or do you rather mean baseMeanLog2? Also, a padj of 14.6 (at the 5th row) is impossible.... And what about the gene identifiers?

So please double-check your input file (ideally you would use the DESeq2 object directly instead of a saved txt file), and when you keep having errors I would recommend you run the ggmaplot function on a data set that everyone can reproduce. If errors still occur, then you know something fundamental is wrong, if that goes fine, it confirms it is specific to your input.

## generate some example data
library(DESeq2)
dds <- makeExampleDESeqDataSet(betaSD=4)
dds <- DESeq(dds)
res <- results(dds)

## Create MA plot

    library(ggpubr)

    ggmaplot(res, main = expression("Group 1" %->% "Group 2"),
             fdr = 0.05, fc = 2, size = 0.4,
             palette = c("#B31B21", "#1465AC", "darkgray"),
             legend = "top", top = 20,
             font.label = c("bold", 11),
             font.legend = "bold",
             font.main = "bold",
             ggtheme = ggplot2::theme_minimal())

enter image description here

ADD REPLY • link 22 months ago Guido Hooiveld ★ 4.1k

0

Entering edit mode

Nathalie ▴ 20

@ab7ad8dd

Last seen 19 months ago

France

Thank you very much for your help and comments. There is errors in my file. BaseMean is BaseMeanLog2. I don't know how insert a file.

ADD COMMENT • link 22 months ago Nathalie ▴ 20

1

Entering edit mode

Yes, some of the values in your input don't make sense.

Yet, since you just started learning below some code to use to read in a tab delimited text file:

# no gene ids included
dtf <- read.table("testinputnoids.txt", header = TRUE)

# with gene ids included in 1st column
dtf2 <- read.table("testinputwithids.txt", header = TRUE, row.names=1)

If you would like to test: these 2 files are available here (for a single download).

ADD REPLY • link 22 months ago Guido Hooiveld ★ 4.1k

1

Entering edit mode

Thank you again for your help. It's very nice of you to take the time to help a newbie. I downloaded your files, it will allow me to better understand my errors. I found my mistakes, I will redo everything calmly. I also have the names of the genes I'm going to use.

ADD REPLY • link 22 months ago Nathalie ▴ 20

score 3 · Accepted Answer · 2023-01-25

3

Entering edit mode

Basti ▴ 780

@7d45153c

Last seen 1 day ago

France

You need to load (and maybe install) ggpubr because ggmaplot is in this package : library(ggpubr) https://www.rdocumentation.org/packages/ggpubr/versions/0.5.0/topics/ggmaplot

ADD COMMENT • link 22 months ago Basti ▴ 780