iiinova — where your digital self keeps evolving

Eric Fauman (he/him) @ericfauman.bsky.social · 19/10/2025

If you didn't get your certificate were you really even there?

Eric Fauman (he/him) @ericfauman.bsky.social · 28/09/2025

I did filter to p < 5e-8. But we also some filtering to independent lead snps. We also have sissies studies ranging from n of a few hundred to a few million subjects.

010

Eric Fauman (he/him) @ericfauman.bsky.social · 28/09/2025

It's a little odd. I am plotting the log10 of the -log10(pvalue) on the x axis so in this plot the null signal drops off much faster than what we observe in the @gwascatalog.bsky.social I'm curious if anyone has a hypothesis why the lead SNP p-values would have this distribution.

110

Eric Fauman (he/him) @ericfauman.bsky.social · 28/09/2025

The effect size in gwas catalog are harder to compare because they are not necessarily in standard deviation units.

100

Eric Fauman (he/him) @ericfauman.bsky.social · 28/09/2025

Oh you win! Yes, all are inappropriate attempts to avoid underflows.

110

Eric Fauman (he/him) @ericfauman.bsky.social · 28/09/2025

Bonus question. Why is the plot of log10(count) as a function of log10(-log10(p)) a straight line? Recall, the data is all the lead SNPs recorded in the GWAS catalog.

020

Eric Fauman (he/him) @ericfauman.bsky.social · 28/09/2025

Quick quiz. This is a plot of how often each p-value appears in this @gwascatalog.bsky.social. Actually I binned the -log10(p) values to their integer values (e.g., 8-9, 9-10, etc). I'm highlighting 3 outliers here, in orange. What are these p-values and why do they diverge so much?

3111

Eric Fauman (he/him) @ericfauman.bsky.social · 28/09/2025

Getting ready for ASHG25 here in Boston! Who's coming to town? As you know, I am fascinated by extreme GWAS p-values. Will we see anything beyond the current record holder, 10^-28538, for eye color. Interestingly, the p-values for all lead SNPs in the @gwascatalog.bsky.social follow a power law

4182

Eric Fauman (he/him) @ericfauman.bsky.social · 16/08/2025

For fun I'm also tracking the room temperature, humidity and CO2 which all increase during the workout, as expected. All the analysis, charting and graphs were done by GPT5 operating on a google sheet where I collected the minute by minute stats.

000

Eric Fauman (he/him) @ericfauman.bsky.social · 16/08/2025

GPT4o pointed me to a device that measures core and skin body temperature. In gold you can see my core body temp continues to rise even after I reach my max heart rate (red), thus requiring further decreases in the work output (blue). Skin temperature drops during the workout, thanks to fans + sweat

100

Eric Fauman (he/him) @ericfauman.bsky.social · 16/08/2025

This graph plots a study I did on myself which involves trying to maintain a heart rate of exactly 123 bpm for 45 minutes. Doing so requires adjusting the work output. In this plot the red line is my heart rate while the blue line is the work output, which you can see decreases over time.

200

Eric Fauman (he/him) @ericfauman.bsky.social · 16/08/2025

Specifically, the phenomenon is called cardiovascular drift and it is a consequence of the body having to do the mechanical work and then also additional work to dissipate the heat generated by the effort, especially the vasodilation to direct internal body heat to the surface.

110

Eric Fauman (he/him) @ericfauman.bsky.social · 16/08/2025

Years ago I asked on Twitter why my heart rate would continue to climb even after I fixed the speed and incline on my treadmill. At the time, no one on Twitter could give me a satisfactory answer. But now GPT4o was able to answer it and suggested approaches to quantify the phenomenon.

192

Eric Fauman (he/him) @ericfauman.bsky.social · 16/08/2025

Isn't that the beginning of the movie "Brazil"?

120

Reposted by Eric Fauman (he/him)

Aaron Rupar @atrupar.com · 24/06/2025

DeGETTE: Will you commit that neither you or anyone else at HHS will take retaliatory action against the brave scientists at NIH? Should be an easy answer bc it's illegal RFK Jr: I can commit that we are absolutely depoliticizing science at NIH for the first time D: I asked you a simple question

1312238574

Eric Fauman (he/him) @ericfauman.bsky.social · 23/06/2025

It could be like designing an antibody to reduce circulating cystatin C levels in an attempt to improve kidney function...

131

Eric Fauman (he/him) @ericfauman.bsky.social · 23/06/2025

And if that's the case, then we haven't really established a causal link between making people walk faster and seeing them age slower.

120

Eric Fauman (he/him) @ericfauman.bsky.social · 26/05/2025

Thanks for these caveats. I guess I'm just surprised a major chromosomal rearrangement can fix in a population so quickly with no apparent impact on the effective population size. Is there good literature on this across other species.

110

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

(I did hedge a bit in my opening post calling it a "possible bottleneck")...

010

Reposted by Eric Fauman (he/him)

Yang Luo @y-luo.bsky.social · 25/05/2025

Pirruccello #eshg2025 - “There is really one game in town right now that is @ukb UKB.” - no one else is producing image data in healthy heart at this scale. @ukbiobank.bsky.social Taking a deep-dive into image-derived phenotypes such as sphericity and cardiac age acceleration.

172

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

This looks great. Reminds me to set aside time to read the whole thing.

120

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

Yes, I saw that paper but don't know enough to weigh the claims of either paper. Have you registered an opinion? We do know chimps have 48 chromosomes and we have 46 so there must have been a first fusion carrier and that must have caused some reproductive difficulties.

120

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

Interestingly the original paper doesn't say the 99% went extinct, it says they were "lost" to our lineage, which is consistent with my model. The quote: "98.7% of human ancestors were lost at the beginning of the bottleneck, thus threatening our ancestors with extinction".

010

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

Any population geneticist want to weigh in on whether any of this makes sense.

310

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

In this model the larger population of unfused individuals persisted for up to 800k years. There's a report of H erectus remains from only 100,000 years ago. Whether we outcompeted our unfused cousins or it was just blind luck I think is not known. www.nature.com/articles/s41...

nature.com

Last appearance of Homo erectus at Ngandong, Java, 117,000–108,000 years ago - Nature

Bayesian modelling of radiometric age estimates provides a robust chronology for Homo erectus at Ngandong (Java), confirming that this site currently represents the last known occurrence of this speci...

120

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

In this model the effective population size reflects some kind of weighted sum of (reduced cross viability with the larger unfused population) and (higher reproductive success with the smaller fusion carriers). In absolute terms I imagine there was a "bottleneck" of one chr2 fusion carrier.

120

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

All the non-fusion carriers are still around but we'd expect reduced reproductive success between fusion carriers and non-carriers. No massive extinction event. Just a genetic barrier between the original population and the fusion carriers.

110

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

Let's imagine a child was born 900kya with a chromosome 2 fusion. Chromosomal fusions occur today in 1:1000 live births. Beating expectations this individual was able to reproduce, nucleating a small cohort of chr2 fusion carriers molecularcytogenetics.biomedcentral.com/articles/10....

120

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

This figure and the coverage of the paper interpret the findings to mean that 99% of the human ancestors alive at the time were wiped out by some cataclysm like major climate change. See here: www.nature.com/articles/d41... But what if there was no extinction event, just reproductive isolation

141

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

This paper from late 2023 got a lot of attention in both the scientific and popular press based on the claim that the effective population size of humans was reduced to 1,280 individuals about 900k years ago www.science.org/doi/10.1126/... This is the authors' model:

140

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

I was today years old when I realized that the Chromosome 2 fusion event which is unique to modern humans probably coincides with a possible population bottleneck event ~900kya. Could the one have caused the other? This is my adaptation of the published figure DOI: 10.1126/science.abq7487

4245

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

In the comments section I saw someone else who got it in 3 and their first two guesses were: CLEAR RIGHT which I thought was almost too perfect.

000

Eric Fauman (he/him) @ericfauman.bsky.social · 25/05/2025

From Adam Butterworth: worth getting embroidered:

151

Eric Fauman (he/him) @ericfauman.bsky.social · 11/05/2025

Happy Caturday! These are Mage and Moxie, brothers to my avatar cat, Mojo.

040

Eric Fauman (he/him) @ericfauman.bsky.social · 02/05/2025

How do you mean? They do a good job of keeping up with published GWAS. And then open targets calculates credible sets for all the full summary statistics in the catalog.

120

Eric Fauman (he/him) @ericfauman.bsky.social · 17/04/2025

Yes, Mojo, the all black cat, is my mews and inspiration. He's patiently waiting for the first pet association wide study, aka PAWS. Also, I'll be at Hinxton from April 28 to May 1st. Love to catch up with anyone who might be around then.

030

Reposted by Eric Fauman (he/him)

ACLU of Massachusetts @aclum.bsky.social · 14/04/2025

We’re in U.S. District Court in Burlington, Vermont for Rümeysa Öztürk’s hearing. Today, the legal team is asking the judge to release her from detention. Ideas aren’t illegal. Rümeysa must be released.

211251313

Eric Fauman (he/him) @ericfauman.bsky.social · 14/04/2025

But do you put two spaces after a period. I hear we're not supposed to do that anymore.

120

Eric Fauman (he/him) @ericfauman.bsky.social · 13/04/2025

It is now 2025. There are 10s of thousands of gwas. I hope @opentargets and similar tools will make it easier for everyone to consider the full complement of phenotypes caused by the genetic variant of interest. In my experience this makes it far easier to infer the likely causal gene.

0120

Eric Fauman (he/him) @ericfauman.bsky.social · 13/04/2025

The two themes, glucose biology and hemoglobin biology, together explain 739 of the HbA1c implicated genes even though they overlap at only 117 implicated genes. I guess HbA1c really is the perfect blend of a metabolite QTL and a protein QTL!

130

Eric Fauman (he/him) @ericfauman.bsky.social · 13/04/2025

The 2nd group includes phenotypes like red blood cell count, bilirubin measurement, reticulocyte count. The 421 L2G associated genes are enriched in red blood cell markers and phenotype maayanlab.cloud/Enrichr/enri...

maayanlab.cloud

Enrichr

120

Eric Fauman (he/him) @ericfauman.bsky.social · 13/04/2025

I'll give the punchline first: Clustering all the phenotypes based on the associated loci reveals two distinct sets of phenotypes The 1st includes traits like: Type 2 diabetes, drugs for t2d, glucose. L2G links 201 unique genes. These are enriched for islet biology: maayanlab.cloud/Enrichr/enri...

maayanlab.cloud

Enrichr

120

Eric Fauman (he/him) @ericfauman.bsky.social · 13/04/2025

So that's what I did. I wrote a script to loop through all HbA1c loci, and for each locus I recorded all the other phenotypes that colocalize with HbA1c. 1300 unique traits 999 unique gene symbols. What did I find?

220