I am very new to the field.
I have a new dataset , GRO-seq from GEO site, and it only consist of raw files (both positive and negative strand measured at different time) in bw format and bed format. I am not sure how to process/annotate/quantify the dataset. I want to transform the data such that it would be a r object, one column for probe id, and a column for expression level. I've heard that I'll need to use rtracklayer, but I'm not exactly sure what I'm suppose to do. And what is the process called, transforming from the raw dataset to gene expression.