Lightnews — Scholar-powered news

Jeff Spence

@jeffspence.github.io

980 followers 730 following 86 posts

assistant professor at ucsf interested in genetics, statistics, etc… jeffspence.github.io

Posts Media Videos Starter Packs

Pinned

Jeff Spence @jeffspence.github.io · Dec 17

What do GWAS and rare variant burden tests discover, and why?

Do these studies find the most IMPORTANT genes? If not, how DO they rank genes?

Here we present a surprising result: these studies actually test for SPECIFICITY! A 🧵on what this means... (🧪🧬)

www.biorxiv.org/content/10.1...

Specificity, length, and luck: How genes are prioritized by rare and common variant association studies

Standard genome-wide association studies (GWAS) and rare variant burden tests are essential tools for identifying trait-relevant genes. Although these methods are conceptually similar, we show by anal...

www.biorxiv.org

4 95 210

Reposted by Jeff Spence

Pragati Kore @pragskore.bsky.social · 2d

📃 We’re excited to share our latest work, now published in Nature Communications — a major update to the Genome Aggregation Database (gnomAD) that improves allele frequency resolution for two gnomAD-defined genetic ancestry groups using local ancestry inference (LAI).

Improved allele frequencies in gnomAD through local ancestry inference - Nature Communications

This study incorporates local ancestry into the Genome Aggregation Database (gnomAD) to improve allele frequency estimates for admixed populations, enhancing variant interpretation and enabling more accurate and equitable genomic research and clinical care.

www.nature.com

1 6 22

Jeff Spence @jeffspence.github.io · 1d

This was such a fun read 🤯🤯🤯

1 2

Reposted by Jeff Spence

Nicholas Mancuso @nmancuso.bsky.social · 1d

This paper is bananas and shows there is still so much to be learned about gradient descent behavior. [ODEs are so hot right now]

centralflows.github.io

Understanding Optimization in Deep Learning with Central Flows

centralflows.github.io

2 1 11

Reposted by Jeff Spence

Struan Grant @struangrant.bsky.social · 6d

My lab is seeking to co-hire a postdoctoral fellow with the
@shaonsengupta.bsky.social team. See details below!

I'll be #ASHG2025, so open to discuss further there if interest.

careers.chop.edu/us/en/job/10...

Post Doctoral Research Fellow – Circadian Biology and Genomics in Philadelphia, Pennsylvania, United States of America | Fellows, Students & Interns at Children’s Hospital of Philadelphia

Apply for Post Doctoral Research Fellow – Circadian Biology and Genomics job with Children’s Hospital of Philadelphia in Philadelphia, Pennsylvania, United States of America. Fellows, Students & Inter...

careers.chop.edu

1 5

Jeff Spence @jeffspence.github.io · 6d

That’s bananas

1 3

Reposted by Jeff Spence

Dr. Arun Sethuraman @arunsethuraman.bsky.social · 8d

Do check out our new manuscript, now out in G3 on modelling theoretical expectations of ghost introgression in sampled genomes. Just in time for Halloween! 🎃 👻

Genetics Society of America @genetics-gsa.bsky.social · 8d

New in #G3journal: @arunsethuraman.bsky.social et al. describe biases in summary statistics, such as genetic differentiation, diversity, Tajima’s D, and estimates of theta, migration rates, and divergence for understanding genomic signatures from extinct ghost populations. buff.ly/7fGnPXj

4 17

Reposted by Jeff Spence

Harold Pimentel @hjp.bsky.social · 9d

Super excited to get this out. This collab started a few years ago and is the first paper from it. Here, with experimental and computational approaches we:

1. establish that cell villages can be just as accurate (one might argue more accurate!) than arrayed-based designs

bsky.app/profile/bior...

bioRxiv Genetics @biorxiv-genetic.bsky.social · 9d

Cell villages and Dirichlet modeling map human cell fitness genetics https://www.biorxiv.org/content/10.1101/2025.09.26.678880v1

2 21 40

Reposted by Jeff Spence

bioRxiv Genetics @biorxiv-genetic.bsky.social · 11d

Fine-tuning sequence to function deep learning models on large-scale proteomic data improves the accuracy of variant effect prediction https://www.biorxiv.org/content/10.1101/2025.09.26.678908v1

1 3

Jeff Spence @jeffspence.github.io · 12d

This should be the “Methods number” for a paper! Like an “Erdős number” or a “Bacon number”. Some papers might have lower Erdős numbers than Methods numbers!

Reposted by Jeff Spence

Tami Gjorgjieva @tamigj.bsky.social · 13d

New paper alert 🚨 w/ @NoahRosenberg, out in EJHG (www.nature.com/articles/s41...) 🧵

Toward minimal SNP sets for record-matching with CODIS STR profiles - European Journal of Human Genetics

European Journal of Human Genetics - Toward minimal SNP sets for record-matching with CODIS STR profiles

www.nature.com

1 1 4

Reposted by Jeff Spence

Harmit Singh Malik @harmitmalik.bsky.social · 19d

Any academic folks on H1B visas (even with stamps in passports) please get legal advice from your University attorneys before leaving the US.

8 140 290

Reposted by Jeff Spence

Hank Greely @hankgreely.bsky.social · 19d

/2 OUTRAGEOUS! Fri. afternoon (night in most of the US), providing less than 25 hr notice of a BIG new fee. Looks like it is not just new post-docs but those who are now overseas. (Note—our academic year starts Monday.) "If you have an H-1B visa & are outside the U.S., you must return immediately"

1 1 2

Reposted by Jeff Spence

Hank Greely @hankgreely.bsky.social · 19d

OH NO!
"The White House issued a proclamation today, Sept. 19, 2025, around 4:36 PM PDT imposing a $100,000 fee for each H-1B visa holder who enters the U.S. after Sept. 21, 2025, 12:01 AM EDT. In CA, this will take effect tomorrow, Sept. 20, 2025, at 9:01 PM PDT." Emergency notice at Stanford.

1 6 6

Reposted by Jeff Spence

gokcen.bsky.social @gokcen.bsky.social · 19d

My team has a postdoc position available now, join us!

careers.gene.com/us/en/job/20...

3 1

Reposted by Jeff Spence

Jesse Engreitz @jengreitz.bsky.social · 20d

Excited for a major milestone in our efforts to map enhancers and interpret variants in the human genome:

The E2G Portal! e2g.stanford.edu

This collates our predictions of enhancer-gene regulatory interactions across >1,600 cell types and tissues.

Uses cases 👇

1/

2 36 84

Reposted by Jeff Spence

Anshul Kundaje @anshulkundaje.bsky.social · 21d

New faculty position opening at @stanford-chemh.bsky.social . Especially interested in candidates with deep expertise and vision for the future of AI/ML/computational models of biomolecules. Please RT

Polly Fordyce @pollyfordyce.bsky.social · 21d

Please apply to our tenure-track faculty position at
@stanford-chemh.bsky.social! We are searching for a new colleague working at the interface between computation and molecular sciences. See post below and pls forward widely!
chemh.stanford.edu/opportunitie...

Faculty Recruitment

chemh.stanford.edu

8 15

Jeff Spence @jeffspence.github.io · 23d

Very cool! Do you have a link to the paper?

Reposted by Jeff Spence

Yun S. Song @yun-s-song.bsky.social · 28d

SINGER, our ARG inference method, is finally published and freely available online:

doi.org/10.1038/s415...

It was a long journey – 16 months from initial submission to acceptance. Is it just me, or has peer review gotten more arduous lately? 4+ rounds of review isn't so unusual these days...

Robust and accurate Bayesian inference of genome-wide genealogies for hundreds of genomes - Nature Genetics

SINGER is a method for creating ancestral recombination graphs to understand the genealogical history of genomes. The method has increased speed, and thus scalability, without sacrificing accuracy.

doi.org

1 51 97

Reposted by Jeff Spence

Joshua G. Schraiber @jgschraiber.bsky.social · 28d

SINGER is a fantastic method for ARG Inference and based on some of the analyses I've seen definitely the best trade off between accuracy and sample size

Yun S. Song @yun-s-song.bsky.social · 28d

Robust and accurate Bayesian inference of genome-wide genealogies for hundreds of genomes - Nature Genetics

SINGER is a method for creating ancestral recombination graphs to understand the genealogical history of genomes. The method has increased speed, and thus scalability, without sacrificing accuracy.

doi.org

1 8 22

Reposted by Jeff Spence

Institute for Human Genetics at UCSF @ihgatucsf.bsky.social · 28d

We invite you to join our seminar at UCSF Mission Bay campus with Brieuc Lehmann from @ucl.ac.uk
humangenetics.ucsf.edu/ihg-seminar-...

1 2

Reposted by Jeff Spence

Lino Ferreira @linoafferreira.bsky.social · 28d

Delighted to see this paper with @anaignatieva.bsky.social now published in Genetics!
academic.oup.com/genetics/adv...

We tackle a thorny issue arising in statistical tests for genetic interactions (epistasis) using ancestral recombination graphs (ARGs)... 🧵

Phantom epistasis through the lens of genealogies

Abstract. Phantom epistasis arises when, in the course of testing for gene-by-gene interactions, the omission of a causal variant with a purely additive ef

academic.oup.com

2 13 21

Reposted by Jeff Spence

Jeremy Berg @jeremymberg.bsky.social · 29d

Here is a word cloud of the words that were missing from the noncompetitive renewal awards but had been present in the parent award.

3/4

A word cloud with dominant words "equity", "disparity", "diverse", and similar words.

12 140 340

Jeff Spence @jeffspence.github.io · Sep 7

They show up in population genetics! Asger Hobolth has a series of nice papers in this space.

1 1

Jeff Spence @jeffspence.github.io · Sep 5

I wonder if this implies that there's not a lot of interaction among the processes and so the actual mutation spectra really are just determined by linear combinations of how active each of these processes are, matching the NMF model assumptions.

1 1

Jeff Spence @jeffspence.github.io · Sep 5

Beautiful work!

It's a bit surprising to me in a statistical sense that so many of the mutation signatures -- learned from NMF -- seem to correspond to actual biological processes.

In contrast it would usually be a fool's errand to find "real" source pop'ns corresponding to pop'ns from ADMIXTURE.

1 2