edgeR question from a inexperienced R user

Question

[Janey]

Hi

Quick start in page 8 of the edgeR User's Guide says:

> group <- factor(c(1,1,2,2))
> y <- DGEList(counts=x,group=group)

I could not change this section of commands according to my data. In fact, compare two data columns together.

my data:

head(x)


                                                   quant4.sf.results                      quant6.sf.results


c51581_g1_i1                                        4.941                                0.000


c48824_g1_i14                                     18.235                               7.628


c19380_g1_i1                                        0.000                                 9.000

Thanks so much for your help.

[Benn]

Hi Janey,

I am not sure what your asking, please take more time to explain exactly what you are doing and what the problem is. Maybe you can explain what your experimental design and research question is, or what errors you encounter. I mean I don't even understand what x is in your data. Are this your counts? If yes, I see the first problem already, counts should be integers. Furthermore, only two samples? n = 1 for each group? :/

[Janey]

Hi b.nota

matrix4_6.txt is my data in gene expression matrices obtained from salmon and trinity software.

                                                                quant4.sf.results                       quant6.sf.results

c51581_g1_i1
4.941 0.000

c48824_g1_i14 18.235 7.628

c19380_g1_i1 0.000 9.000

One column is control and other column is treatments.

X was obtained from this command:

x <- read.delim("matrix4_6.txt",row.names="Symbol")

group <- factor(c(1,1,2,2))

y <- DGEList(counts=x,group=group)

In fact, I do not know after first command, I have to do what??

[Benn]

I am not sure what your understanding of statistics is, but groups with n = 1 are not sound statistics. In the edgeR manual there is even a small section about n = 1 experiments, did you read that?

Furthermore, in the same manual, they explain that you need read count data, which you don't have. Counts are integers (whole numbers without decimals), did you read that part of the manual?

My advice is to read the manual first.

[Janey]

Hi b.nota

Thank you for helping me to solve my problem.