Daniel E. Weeks

Please RT!

🧬Postdoc Opportunity in Statistical/Computational Genetics!🧬 Join our team in Human Genetics @PittPubHealth! Focus: Genetic etiology of cardiometabolic disease in Samoans & Pacific Islanders. Collaboration with @CharlestonCWKC tinyurl.com/pitthugen #Postdoc #Academicjobs

Feeling frustrated when trying to put together your dbGaP data submission? @StatGenDan and I are excited to announce our new R package, #dbGaPCheckup, which we designed to help YOU with this complex process! 🧵(1/n) #Rstats #Genetics #dbGaP #DataQC #DataSharing #OpenScience

@michelnivard Challenging to do (even somewhat) effectively without amplifying. This guide may have some good ideas: @brendannyhan/109546356575773277" target="_blank" rel="nofollow noopener">mastodon.social/@brendannyhan/…

@OcOutlier I hope it all goes smoothly, proves very helpful, and that your recovery is smooth and swift.

Tomorrow I am having major, experimental, spinal surgery. Please send your prayers and well wishes as my family and I need them!

We are excited to share our new work @Nature, led by @Lenka_Dohnal, uncovering a gut-brain pathway that regulates the motivation for #exercise. (1/7) nature.com/articles/s4158…

@emollick The made-up incorrect “facts” & frequent wrong definitions in the midst of convincing sounding text markedly reduces the value of ChatGPT output & creates the need for extensive and accurate fact-checking. Always wondering which parts of its output are wrong. It is often wrong.

The quality of the initial essay is, as it often is, rather astonishing. But it also suffers from the three main weaknesses of AI writing on complex topics: 1. Vagueness and lack of clear examples 2. Oversimplification/missing nuance 3. Made-up facts Let's fix some of those. 3/

A killer app of AI is its ability to multiply an expert's work. So, a 🧵of tips to use ChatGPT to power up your writing, assuming you are an expert on the topic. Lets pretend we were trying to write my post on how AI can make you creative, using AI. 1/ oneusefulthing.substack.com/p/how-to-use-a…

We released a commentary on Ioannidis et al (2021) as a preprint. To date the settling of Polynesian islands, they inferred split times using IBD segments shared between modern genomes. We show that their estimator is biased to be ca. 300 years too old. doi.org/10.1101/2022.1…

@SundermannAJ I see two accounts for you on Mastodon. The second is @SundermannAJ@mas.to ?

Bravo "Annals of the Rheumatic Diseases" for giving the option to download a combined PDF containing both the manuscript and the Supplemental Material! This is a time saver, as I almost always end up needing to read something in the Supplemental Material.

"it’s also wise to simply not use filenames that contain spaces or other strange characters. Simple alphanumeric names with either dashes or underscores are best." @vsbuffalo in his book "Bioinformatics Data Skills".

@doctorveera Wonderful thread! However, the MAF is not 0.025 (The map fig legend is confusing). All of the MAF estimates were <0.025 with wide confidence intervals, and in our largest sample, from Samoa, the MAF was 1.25% [95% CI 0.88%- 1.75%]. Fig here based on Table 2. Need more data.

This variant is almost absent in most of the world but has a MAF of 0.025 in Western Polynesia. Though it's tempting to assume that there might be a beneficial effect that resulted in positive selection, the authors write the high MAF is likely due to genetic drift.

Genetic discoveries such as this are mind-blowing not only because they are absolutely fascinating--both clinically and biologically--but they also demand immediate attention from the policy makers. 🧵 twitter.com/casanova_lab/s…

@TonyMerriman2 Nice thread! However, not sure the allele frequency is ~2% as all but one of the minor allele frequency estimates were smaller with wide confidence intervals, and in our largest sample, from Samoa, the MAF was 1.25% [95% CI 0.88%- 1.75%]. Fig based on Table 2. Need more data.

'if standardized coefficients do not add information, they certainly do not add meaning. "To replace the unmeasurable by the unmeaningful is not progress" (Achen, 1977, p. 806).' - King G. American Journal of Political Science. 1986 Aug;30(3):666–687.

I have set up a Mastodon account at @StatGenDan@fediscience.org, and used pruvisto.org/debirdify/ to easily populate my follows list. As that works by searching your Twitter profile for your Mastodon ID, please add your Mastodon ID to your profile so I can find you on Mastodon.

This guide to Mastodon is great.

Rule 3 revised: If possible, cut words.

Sounds good to me. My call sign is KB3WIO.

@ProfJohnATodd @jcbarret My first genome-wide linkage study in 1988 used 171 RFLP markers! I don’t show up as an author - PubMed couldn’t handle long authorship lists: RA Gatti, I Berkel, E Boder, G Braedt, P Charmley, P Concannon, F Ersoy, T Foroud, NG Jaspers, K Lange, et al pubmed.ncbi.nlm.nih.gov/3200306/

@jcbarret You’re just a kid, Jeff- my first polygene discovery was in 1986. It’s taken all this time to work out mechanism. I feel that we have the ideas and questions but have to wait for technology to catch up. It was, e.g., an over ten yr wait for genome-wide SNP arrays.

A speaker at #ASHG22 just referred to the progress from GWAS “over the past two decades” and I felt pretty old. 👴

@IanJMackay Yes, please do!

@StatGenDan I thought you'd forgotten Dan. I'll send you my updated collection! Though they are a bit more plant focused!

“Mathematics may be compared to a mill of exquisite workmanship which grinds you stuff of any degree of fineness, but nevertheless what you get out depends on what you put in - … so pages of formulae will not get a definite result out of loose data.” TH Huxley