Read on Twitter
ICYMI*, we had a new paper come out in @NatureGenet last week: a GWAS of "noncognitive skills," a construct that nearly everyone can agree is important, but no one can agree on how to measure:
https://rdcu.be/cddNY
(*I, for one, missed everything science-related last week)
https://rdcu.be/cddNY
(*I, for one, missed everything science-related last week)
We got around the problem of how to measure non-cognitive skills by ... not measuring them! Instead we used GWAS-by-subtraction to isolate genetic variation in educational attainment that was *not* shared with cognitive ability.
Here's a tutorial: https://rpubs.com/MichelNivard/565885
Here's a tutorial: https://rpubs.com/MichelNivard/565885
So, a very literal definition of non-cognitive skills -- anything that is not the cognitive abilities measured by the IQ tests used in the GWAS of cognitive ability. Non-cognitive skills is a phrase that many folks (myself included) loathe, but this approach has a big advantage:
It allows us to take the SNPs associated with the latent (i.e., unmeasured) non-cognitive skills and use them as "genotypes in search of phenotypes" -- what sorts of things that *have* been measured are correlated with these genes?
Not surprisingly, loads of stuff!
Not surprisingly, loads of stuff!
Not everything correlated with non-cognitive genetics is "good" -- in particular, we see positive genetic correlations w/ some forms of mental illness
We summarize our main findings, give a crash course in methods, and describe key caveats in this FAQ: https://medium.com/@kph3k/investigating-the-genetic-architecture-of-non-cognitive-skills-using-gwas-by-subtraction-b8743773ce44
We summarize our main findings, give a crash course in methods, and describe key caveats in this FAQ: https://medium.com/@kph3k/investigating-the-genetic-architecture-of-non-cognitive-skills-using-gwas-by-subtraction-b8743773ce44
The code for our analyses can be found here: https://github.com/PerlineDemange/non-cognitive
The input summary statistics are from @GWASCatalog:
ftp://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics//GCST90011874/
ftp://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics//GCST90011875/
The input summary statistics are from @GWASCatalog:
ftp://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics//GCST90011874/
ftp://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics//GCST90011875/
This paper is @PerlineDemange 's first paper (NBD, just a first-authored paper in Nature Genetics that came out on her birthday)! Also @MarghMalanchini @wakeworksleep @pietrobiroli @tuckerdrob @Danbelsky @michelnivard + others -- a really fun team
What I personally am hoping this work will be useful for (a partial and speculative list):
- clarifying the sometimes perplexing pattern of genetic associations that you see w/ educational attainment
- refining theory re: most important constituents of non-cognitive skills...
- clarifying the sometimes perplexing pattern of genetic associations that you see w/ educational attainment
- refining theory re: most important constituents of non-cognitive skills...
- underscoring YET AGAIN that genes associated with educational attainment are not "good" genes
- serving as a tool for researchers aiming to understand heterogeneity in response to educational interventions, esp. as many target noncognitive skills
- serving as a tool for researchers aiming to understand heterogeneity in response to educational interventions, esp. as many target noncognitive skills
- helping parse what, exactly, is being transmitted in indirect genetic effects / genetic nurture
- giving researchers a tool to trace the heterotypic continuity of personality development over the lifespan
(am probably missing heaps of other interesting applications)
- giving researchers a tool to trace the heterotypic continuity of personality development over the lifespan
(am probably missing heaps of other interesting applications)
Stay tuned for follow-up work from @PerlineDemange and others ... She is definitely a rising star to watch!
