There are still important details lacking. 1) How to address the substantial differences in prediction calibration and uncertainty across different environmental contexts (see: Hou et al. 2024 pubmed.ncbi.nlm.nih.gov/38886587/), many of which are unknown for an embryo by definition?

Very nice 🔥are you aware if syllabous/ material is available somewhere? I've been thinking about a similar course for PhD students, but it'd be great to know of an already-tested course

I am starting to give up on taking motivation letters as a way to learn more about the applicant 😔 I think truly good candidates are drafting very good letters, but then we might be missing out on many other good candidates that are crafting impersonal cover letters by using LLM in excess.

Yeah, but even in GWAS I'd expect != genes prioritized, downstream analyses, etc. But it's true that there's an ocean of methods/ QC in scRNA studies. Maybe it would be a good start to standardizing pipelines? If that is anything to be seeked in the field

A couple years ago there was an initiative similar to what you mention for GWAS and downstream analyses. They would provide a question + data to independent groups. I couldn't find the project website any longer, but it's likely out there.

The fake ID is the same across all databases one asks access to, but are project specific. Authorities keep linkage file (fake ID to personal ID) which can authority uses to retreive additional data (if needed). Pretty robust, but data is considered reidentifiable anyway (at least in Sweden).

I will chip in just to say that the data is likely pseudonomyzed (as is the case in Sweden), where personal number given at birth is stored by national authorities and they do the linkage for you (or exctract the data you require). We only get a fake id that connects multiple registers/ data.