Florian Privé

icon indicating a website link https://privefl.github.io/ icon indicating an email [email protected]

icon company National Center for Register-based Research (NCRR, Aarhus University) icon location Aarhus, Denmark // Lyon, France icon location Senior Researcher (2022–) in Statistical Human Genetics • Postdoc (2019–2021) • PhD student (2016–2019) • ENSIMAG (2013–2016)

Results 380 comments of Florian Privé

fix rsid-based `snp_asGeneticPos()`

Thanks for this investigation. If I understand correctly, the problem arises because the genetic positions are not monotonic with the hg38 positions. Then the question is: do we really want...

fix rsid-based `snp_asGeneticPos()`

The thing with kNN is that it is okay because the data is so dense in the genetic map, which might not be the case when considering only the variants...

fix rsid-based `snp_asGeneticPos()`

What about using `method = "fmm"`? It does not need monotonic positions.

fix rsid-based `snp_asGeneticPos()`

Yes, please tell me if results with this look reasonable.

fix rsid-based `snp_asGeneticPos()`

Thanks for reporting. I'll try when I have some time this week.

fix rsid-based `snp_asGeneticPos()`

I'll try to look at this again next week.

fix rsid-based `snp_asGeneticPos()`

So, I think there are two okay solutions: 1. Use a linear interpolation: `approx(pos.chr, new_pos, xout = pos.chr[indNA], rule = 2)$y` 2. Do not try to interpolate positions corresponding to...

Add cross-trait LDSC

Thanks for this! I'll look at it next week. Could you point me to some resources you used to base your implementation on?

Add cross-trait LDSC

Thanks for the refs. Yes, there is still quite some work to be done.

Add cross-trait LDSC

I agree it would be great to have this in the package. However, there is still quite some work to be done to finish implementing this. And I have absolutely...