Hi there,
I wish to run analysis of counts from a published RNAseq experiment to look for genes that are differentially expressed in a knockdown/knockout cell line, with the aim to identify common transcripts between these data and data produced from my lab. However, I'm unable to match the sample IDs (control/Knockdown) with that of the GEO Accession IDs/SRA runs as the column headers for each run are listed rather ambiguously (e.g. v11, v12, v13 etc.). The dataset I'm looking at is listed here:
https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE124407 GSE124407_RNAseq_count_all_matrix_siRNA.txt.gz GEO Accessions: GSM3532120-25
I'm able to run the analysis no problem with DESeq in R and get the genes I need but at the moment its a bit of an educated guess as to which samples the GEOs represent, as I know that the knockout will amplify expression of a gene it normally represses. As such, I'd be grateful for any help or advice on to how to get around this?
Thanks in advance!