Firstly you should read the paper more than 3 times :
PLoS One. 2014 Jun 13;9(6):e99625. doi: 10.1371/journal.pone.0099625. eCollection 2014.
RNA-Seq transcriptome profiling identifies CRISPLD2 as a glucocorticoid responsive gene that modulates cytokine function in airway smooth muscle cells.
In that paper, we can find the raw data of this experiment, which is located in : https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE52778
If you are familar with RNA-seq workflow , you can process those data from begin to end, get the expression matrix by your self, but you can also load the expression matrix by R package: airway
The readme for this package: https://bioconductor.org/packages/release/data/experiment/vignettes/airway/inst/doc/airway.html
As we can see, the samples in same group are more similar with each other than samples in different group.
It's very easy to understand, so I don't want to explain the codes one by one.
You can also read my tutorial rnaseq-workflow.txt to study how to process the raw fastq data of RNA-seq.
The R code:
step0-install.R
step1-check.R
step2-DEG.R
step3-go-kegg.R
It's a little complicate for kegg_and_go_up_and_down.R and run_DEG_RNA-seq.R, bless you.