@DianaCubas18 @genepisociety Hi Diana—can you DM me with details on your experience, eg any error messages you’ve received. Others have been able to register. Let’s get you signed up too.
I spend days every year talking with multiple IRBs trying to sort out when coded data are or are not “human subjects research” and are or are not subject to the single IRB expectation—and meanwhile the Supreme Court has decided COVID is not a workplace issue. Disheartening.
Sometimes, during my daily meditation, I sit with this fact, in curiosity and wonder: I, also, have had absolutely no reverberation.
Happy to announce that a project @GenEpi_Sara and I have been working on for over five years has passed the penultimate point on this curve... (HT Hugues Aschard)
Broad street businesses were complaining so I reinstalled the pump handle.
For those interested in “multi-phenotype analysis in human genetic studies,” check out this talk (and Hugues Aschard’s work generally). twitter.com/CANSSIOntario/…
New Faculty Position: We invite applications from candidates for a faculty position as assistant or associate professor of epidemiology, with a focus on Alzheimer's disease and other sources of age-related cognitive decline. Apply here: ow.ly/UR2M50HelRn @HarvardChanSPH
@paulpharoah @sparkys_mom Sure—which hints at a deeper point re: contextuality of art. Which might be relevant to aspects of the controversy around the painting: it’s not “what is this depicting” but “what is it DOing.”
@EpiEllie Giving off some real “Disclosure” (1994) vibes on how we’ll manage our digital files virtually.
@priya4genes @FallinDani @genandgenes @GnomeChris Similarly "Program in Molecular and Genetic Epidemiology" --> "Program in Genetic Epidemiology and Statistical Genetics" reflects our work and encompasses multiple investigators with overlapping interests & trainees.
Want to learn about GWAS, its QC, analysis tools, imputation, meta-analysis, PRS and MR? Apply for our 7th edition of online LIVE 5-day course “Intro to statistical analysis of GWAS”! @UniOfSurrey with support from @eshgsociety tinyurl.com/2p8wurv6 Please, RT
📢HIRING: Open positions for postdocs interested in better understanding the role of genetic ancestry and population-specific epidemiology on genetic risk in admixed populations with me at @JohnsHopkinsEPI.
Please share and let me know if interested!
Human genetics studies are motivated by a desire to understand disease and eventually help people by improving prediction, prevention and treatment of disease. A 🧵and a 📰
News&Views here: nature.com/articles/d4158…
Please join us for the next International Genetic Epidemiology Society Journal Club on 13 December 9 am EST: Henrike Heyne will lead the discussion of "Mono- and bi-allelic effects of coding variants on disease in 176,899 Finns." Register here: harvard.zoom.us/meeting/regist…
I want to take a moment to talk about a Motte and Bailey tactic that often gets disingenuously used to frame and restrict conversations around human differences among groups. 1/n
@twitskeptic @cecilejanssens Further, as noted in the cited article [Elliott JAMA 2020] after adding PRS, estimates of risk “changed by 5% or more for 1.1% of participants.” Given one recommended risk threshold for lipid lowering meds is 7.5%, that’s a meaningful change for those in that 1.1%.
@twitskeptic @cecilejanssens To give some sense of scale and highlight the fact that the C-statistic is not a direct indicator of clinical utility: If I’ve done the maths right, the increment in AUC comparing population BRCA1 testing versus not is about .02.
@cecilejanssens @timfrayling @michelnivard I do not think this is empirically true for most complex diseases. The per-PRS-sd change in log odds is roughly the same in models with or without risk factors, even though some of the PRS effect is mediated thru risk factors (which are non-exhaustive and measured with error).
@cecilejanssens Readers can disagree whether we currently know enough about a particular setting to determine whether PRS are worth adopting, and they may have doubts about future utility. But there are some settings I would not dismiss as “laughable”—I count CAD among those. [2/2]
@cecilejanssens The framework the authors lay out for possible use cases and what needs to be evaluated, what individual and societal risks need to be ameliorated for PRS to be deemed beneficial is sensible. [1/2]
@cecilejanssens Noted by the authors as the first research gap to be filled.