DESeq2 analysis with R

Question

[z.ghseminezhad92]

Hi, I have 19 Salmon.quant sample(9 control,10 cancer)with csv format that merge them with tximport and now want to perform DESeq2 analysis for my merged data to find what transcripts has diffrent expression in cancer and control my sample_key is like this:

NO.       group           sample 
1            control          GSM366.
2            control          GSM366.
3            control          GSM366.
4            cancer          GSM366.
5            cancer          GSM366.

my txi.salmon is like this:

                        abundance.GSM366.csv....    counts.GSM366.csv    lenght.GSM366.csv
transcript1           0                                                   0                                3.25 
transcript2           4.61                                            0.5                               3.85
transcript3            0                                                  0                                3.26
transcript4           10                                                 1                                3.54
....

and these are my R script:

>library(tximport)
>library(readr)
>setwd("G:/Palindrome/salmon_OUT4/")
>SalmonQuantdir<-"G:/Palindrome/salmon_OUT4/Files"
>setwd(SalmonQuantdir)
>files<-list.files(SalmonQuantdir)
>names(files)<-files
>tx2gene<-read.csv("G:/Palindrome/salmon_OUT4/tx2gene.csv")
>txi.salmon <- tximport(files, type="salmon", tx2gene=tx2gene, importer =read.csv, abundanceCol = "TPM")
>Key_file=read.csv("Sample-key.csv")
>library("DESeq2")

but i don't know how to design condition and how to creat countdata and then perform DESeq2. please help me if you know what scripts is appropriate for R?

Answer 1

[jared.andrews07]

Reading the DESeq2 vignettes and associated workflow are good places to start. They are quite easy to follow and explain each step in-depth to help you understand not only the how, but the why. Your design would just be design = ~ group.