Seriously? The topics in the second half of the class are what turned R into something notable. People would still be jerking around with Matlab or SAS (and getting poorer results in many cases) if Gentleman hadn't decided that microarrays were kinda interesting.
I'm biased as I was one of the early BioC devs but totally agree. What I saw back then (~2000) was a ridiculously niche language (R) in a fairly niche field (statistics). BioC brought in a ton of new blood by way of the biology folks.
It wasn't really until the Hadleyverse picked up several years later that I started to see non-genomics related mainstream use of R
But mapping and alignment in R? Genome annotation in R? I've worked in 3 institutions with a variety of bioinformaticians and that is something I've never seen. I'd argue that R is an unusual choice for that.
Microarrays yes, but no one (should) really uses those anymore.
Don't get me wrong, I do 50-75% of what I do in R.
Mapping an alignment gets farmed out to C via rsubread, samtools, etc. as far as annotation, anyone who doesn't use GRanges (or vcfanno, bed tools, etc) is insane. (You shouldn't be aligning locally anyways, it's a waste of resources in most cases. The broad migrated their workflow to Google last I heard; we do a lot in BaseSpace.). I actually don't see the point of alignment at all for RNA.
Your data almost certainly has batch effects, you know. At least with arrays the prep was fully standardized. Outside of a few big shops (UBC, the Broad, etc) it's the goddamned Wild West in terms of prep for lots of what we used to do with microarrays, especially RNAseq.
And of course since single cell techniques are newish, nutty ideas like designed experiments are only just beginning to dawn on people. Fisher would be apoplectic.
"An introduction to basic statistical concepts and R programming skills necessary for analyzing data in the life sciences."
https://www.edx.org/course/statistics-r-harvardx-ph525-1x