Jeff Spence

Jeff Spence

@jeffspence.github.io

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

1.2K followers918 following210 posts

Jeff Spence reposted

Willow Coyote-Maestas
@willowcoyote.bsky.social

We will be hosting 4th Full Circle Symposium for Native Biologists hosted at UCSF in May 2027! We aim to build community and provide meaningful opportunities for Native scientists. If you are a Native biologist at any training stage who wants to speak, plz fill this out by this 1/31: bit.ly/4kh5GFx

Jeff Spence reposted

Adam MacLean
@adamlmaclean.bsky.social

We are hiring a tenure-track assistant professor in computational biology at USC! area: AI method development and its applications in biology. Do you combine quantitative research with a drive to develop innovative approaches for tackling biological questions? usccareers.usc.edu/job/los-ange...

usccareers.usc.eduAssistant Professor of Quantitative and Computational Biology at USCLearn more about applying for Assistant Professor of Quantitative and Computational Biology at USC

Jeff Spence reposted

loic-yengo.bsky.social
@loic-yengo.bsky.social

If you’ve ever run a GWAS, chances are you've used COJO (Conditional and Joint analysis). Well, we've extended and improved (up to 100× faster 🚀) this essential tool in the GWAS toolkit to enable joint analyses across multiple ancestry groups. Paper: www.nature.com/articles/s41...

nature.comMultiancestry conditional and joint analysis applied to GWAS summary statistics - Nature GeneticsThis study introduces Manc-COJO, a method for detecting independent genetic associations by integrating data from multiple ancestries, reducing false positives and enabling fast, flexible analyses acr...

Jeff Spence reposted

Nikhil Milind
@nikhilmilind.dev

I'll be at #ASHG2026 this year! I'm presenting some recent work (with @huishengzhu.bsky.social) on extracting all the information we can out of rare-variant burden test statistics for downstream meta-analysis. Catch me at poster 8357F to talk about the project or stat gen / pop gen.

Jeff Spence reposted

PLOS Biology
@plosbiology.org

The ratio of nonsynonymous to synonymous DNA substitutions (dN/dS) encodes information about selection pressures. This study develops a model of dN/dS in recombining bacterial populations, offering insights on how horizontal gene transfer shapes the genome. 🧪 #AcademicSky #evolution plos.io/3Wcy9lT

Jeff Spence reposted

Jonathan Pritchard
@jkpritch.bsky.social

Delighted to share this new article where we propose our view for the next generation of human genetics. We argue a need for "Trait Causal Models" linking genetics to effects at all levels of biological structures, and talk about how to get there using gwas, perturbations and observational omics

Jeff Spence@jeffspence.github.io

It's been 30 years (!!!) since the conceptualization of GWAS. GWAS has driven tons of discoveries, but with recent experimental developments (CRISPR, single-cell omics), we thought it was time to think about how to drive the next 30 years of human genetics. doi.org/10.5281/zeno...

doi.orgTrait Causal Models: from GWAS, perturbations, and omics to predictive models of human biologyA central goal of human genetics is to understand the molecular pathways that link genes to functions in cells and tissues, and ultimately to their roles in health and disease. Genome-wide association...

Jeff Spence reposted

Emma Dann
@emmamarydann.bsky.social

Stoked to share our new perspective on ✨ Trait Causal Models ✨ as the next frontier for human genomics! It was a pleasure to put this together with @jeffspence.github.io and @jkpritch.bsky.social with guidance from @minetoota.bsky.social @jengreitz.bsky.social @marsonlab.bsky.social Thread 👇

Jeff Spence@jeffspence.github.io

It's been 30 years (!!!) since the conceptualization of GWAS. GWAS has driven tons of discoveries, but with recent experimental developments (CRISPR, single-cell omics), we thought it was time to think about how to drive the next 30 years of human genetics. doi.org/10.5281/zeno...

doi.orgTrait Causal Models: from GWAS, perturbations, and omics to predictive models of human biologyA central goal of human genetics is to understand the molecular pathways that link genes to functions in cells and tissues, and ultimately to their roles in health and disease. Genome-wide association...

Jeff Spence reposted

Blue Eyed Grass
@blueeyedgrass.bsky.social

Enjoyed this. Positive, forward thinking with existing technology. The perturbome, I suppose. And it defines what success means for that-predictive power. That's different from mechanism. But mechanism means different things to different people, and, in any event, can be really hard.

Jeff Spence@jeffspence.github.io

There have been incredible advances across all 3 of these data pillars, but we discuss what more needs to be done to move from these pillars to building TCMs.

Jeff Spence
@jeffspence.github.io

It's been 30 years (!!!) since the conceptualization of GWAS. GWAS has driven tons of discoveries, but with recent experimental developments (CRISPR, single-cell omics), we thought it was time to think about how to drive the next 30 years of human genetics. doi.org/10.5281/zeno...

doi.orgTrait Causal Models: from GWAS, perturbations, and omics to predictive models of human biologyA central goal of human genetics is to understand the molecular pathways that link genes to functions in cells and tissues, and ultimately to their roles in health and disease. Genome-wide association...
View full thread
Jeff Spence
@jeffspence.github.io

As an addendum, AI slop has made it difficult to post perspective pieces on preprint servers, so we're sharing ours on Zenodo. We did use LLMs to catch typos etc..., but rest assured that human minds conceived the work, human fingers typed it up, and human eyeballs proofread it.

Jeff Spence
@jeffspence.github.io

It's been 30 years (!!!) since the conceptualization of GWAS. GWAS has driven tons of discoveries, but with recent experimental developments (CRISPR, single-cell omics), we thought it was time to think about how to drive the next 30 years of human genetics. doi.org/10.5281/zeno...

doi.orgTrait Causal Models: from GWAS, perturbations, and omics to predictive models of human biologyA central goal of human genetics is to understand the molecular pathways that link genes to functions in cells and tissues, and ultimately to their roles in health and disease. Genome-wide association...
View full thread
Jeff Spence
@jeffspence.github.io

We discuss all of this and more in the piece. It was a lot of fun thinking through how to use the data that we have and to try to crystallize what the goal of human genetics should be. We're hoping to spark a conversation so please let us know you're thoughts. doi.org/10.5281/zeno...

doi.orgTrait Causal Models: from GWAS, perturbations, and omics to predictive models of human biologyA central goal of human genetics is to understand the molecular pathways that link genes to functions in cells and tissues, and ultimately to their roles in health and disease. Genome-wide association...
Jeff Spence
@jeffspence.github.io

It's been 30 years (!!!) since the conceptualization of GWAS. GWAS has driven tons of discoveries, but with recent experimental developments (CRISPR, single-cell omics), we thought it was time to think about how to drive the next 30 years of human genetics. doi.org/10.5281/zeno...

doi.orgTrait Causal Models: from GWAS, perturbations, and omics to predictive models of human biologyA central goal of human genetics is to understand the molecular pathways that link genes to functions in cells and tissues, and ultimately to their roles in health and disease. Genome-wide association...
View full thread
Jeff Spence
@jeffspence.github.io

There have been incredible advances across all 3 of these data pillars, but we discuss what more needs to be done to move from these pillars to building TCMs.

Jeff Spence
@jeffspence.github.io

Key challenges include: - Extracting directional, quantitative effects of _genes_ on traits from GWAS data - Choosing cell types and conditions and types of perturbations in screens - Choosing the right level of abstraction for modeling the path from variant --> trait

Jeff Spence
@jeffspence.github.io

It's been 30 years (!!!) since the conceptualization of GWAS. GWAS has driven tons of discoveries, but with recent experimental developments (CRISPR, single-cell omics), we thought it was time to think about how to drive the next 30 years of human genetics. doi.org/10.5281/zeno...

doi.orgTrait Causal Models: from GWAS, perturbations, and omics to predictive models of human biologyA central goal of human genetics is to understand the molecular pathways that link genes to functions in cells and tissues, and ultimately to their roles in health and disease. Genome-wide association...
View full thread
Jeff Spence
@jeffspence.github.io

TCMs should be: - Mechanistic: a model capturing the mechanisms by which genes affect a trait should be able to convert knowledge about how a perturbation or variant affects an intermediate phenotype into predictions about how it would affect the trait.

Jeff Spence
@jeffspence.github.io

To build TCMs will require 3 distinct "data pillars", each with complementary advantages 1. Genetic association data (e.g., GWAS) 2. Perturbations in model systems (e.g., perturb-seq in cell lines) 3. Human subjects omics profiling (e.g., scRNA-seq from cases and controls).

Jeff Spence
@jeffspence.github.io

It's been 30 years (!!!) since the conceptualization of GWAS. GWAS has driven tons of discoveries, but with recent experimental developments (CRISPR, single-cell omics), we thought it was time to think about how to drive the next 30 years of human genetics. doi.org/10.5281/zeno...

doi.orgTrait Causal Models: from GWAS, perturbations, and omics to predictive models of human biologyA central goal of human genetics is to understand the molecular pathways that link genes to functions in cells and tissues, and ultimately to their roles in health and disease. Genome-wide association...
Jeff Spence
@jeffspence.github.io

We argue that the goal of human genetics should be to build *Trait Causal Models*. TCMs are quantitative, predictive models that trace the biological processes from genes --> traits.

Jeff Spence
@jeffspence.github.io

TCMs should be: - Quantitative: we should move beyond just cataloging associations and instead we should estimate _by how much_ and _in which direction_ a variant or a perturbation would affect the trait

Jeff Spence reposted

Matthew Aguirre
@aguirre404.bsky.social

Happy to share some new work done in collaboration with my colleagues in Human Genetics at Genentech! In it, we assess five model-based variant pathogenicity labels in the context of rare variant burden testing. [Summary ⬇️]. link.springer.com/article/10.1...

link.springer.comMarkedly divergent performance of variant annotation methods for gene-level association testing - BMC GenomicsMachine learning-based annotation methods are increasingly used to assess the pathogenicity of genetic variants, but their performance at prioritizing variants for gene-level association testing remai...

Jeff Spence reposted

Débora Y C Brandt
@deboraycb.bsky.social

Now that it's officially out, I have to say I'm quite proud of this work🙂 I got involved in it after a bit of a failure in another proj. Took way longer than it could. Started pregnant, published close to baby's 3rd bday😅 Very grateful for my patient mentors and collaborators doi.org/10.1093/molb...

academic.oup.comCharacterizing the detectable and invisible fractions of genomic loci under balancing selectionAbstract. Balancing selection refers to scenarios where selection maintains genetic polymorphisms affecting fitness and its components. Empirical studies h
Max Reuter@maxreuter.bsky.social

Our paper on detecting (or not) balancing selection is now out, in all its typeset beauty, in the October issue of @molbioevol.bsky.social: academic.oup.com/mbe/article/...

Jeff Spence reposted

Hanbin Lee
@epigenci.bsky.social

Replying to @epigenci.bsky.social

We then applied the theory to human demography and deduced a provisional estimate of stabilizing selection in humans accounting for LD. The key machinery is Roshni Patel's conditional frequency spectra. With SDEs, this becomes a PDE scheme.

Jeff Spence reposted

Hanbin Lee
@epigenci.bsky.social

We studied the evolution of genome-wide linkage disequilibrium and its influence on genetic architecture with genetic drift. We found that the Bulmer effect is population size and demography dependent. For nerds: genetic architecture follows from solving PDEs.

bioRxiv Genetics@biorxiv-genetic.bsky.social

Linkage disequilibrium moderates how evolving genetic background shapes variant-level architecture https://www.biorxiv.org/content/10.64898/2026.10.01.756092v1

Jeff Spence reposted

Roshni Patel
@roshnipatel.bsky.social

So excited about this work with @nasalab.bsky.social! Led by Stephanie Gogarten & with help from @daphmarts.bsky.social, we use LLMs as judges to evaluate the use of population descriptors in the entirety of the GWAS Catalog (~4,000 papers published 2007-2025) www.biorxiv.org/content/10.6...

biorxiv.orgLarge language model-based bibliometric evaluation of population descriptors in human geneticsAs the use of population descriptors such as race, ethnicity, and ancestry have become increasingly common in modern genetics research, there have been growing calls to critically examine their use. M...

Jeff Spence reposted

Tom Booker
@tombooker.bsky.social

I’m keen to share this preprint. There is growing evidence that associative overdominance leads to the maintenance heterozygosity and we wondered how this would affect variance for traits under stabilizing selection: www.biorxiv.org/content/10.6...

biorxiv.orgMaintenance of quantitative genetic variation in regions of restricted recombination via associative overdominanceMutation-selection balance is the most plausible general mechanism for the maintenance of additive genetic variance (VA) for quantitative traits in natural populations. It is well established that the...

Jeff Spence reposted

Moi Expósito-Alonso (MOILAB)
@mexpositoalonso.bsky.social

📣 We are looking for a PhD student in 2027! Interested in studying rapid evolutionary dynamics of species in changing climates? Apply to the MOI LAB by Nov 19! Flyer below & tinyurl.com/flyerphdmoilab @ucberkeleyofficial.bsky.social @innovativegenomics.bsky.social @hhmi-science.bsky.social

tinyurl.com2027_PhD_RapidEvolutionGenomicsEcology-MoiLab_Berkeley-HHMIPhD positions: Rapid evolutionary genomics in plants Moi Exposito-Alonso Lab — www.moilab.science Assistant Professor, Freeman Hrabowski Scholar HHMI · Department of Integrative Biology · Innovative G...

Jeff Spence reposted

Adam Auton
@adamauton.bsky.social

What can we learn about a single rare variant? Published today in Science: the 23andMe Research Institute and Dana-Farber studied the EGFR T790M in 10.1M 23andMe research participants. The variant has OR = 25.2 for lung cancer, rising to 61.7 in never-smokers(!). 🧵

Jeff Spence reposted

Matthew Hahn
@3rdreviewer.bsky.social

New paper with Trang Nguyen! Do you have only a single parent and offspring, but still want to calculate mutation rates? We've got the tool for you--OOPS github.com/TrangNg-Th/O... www.biorxiv.org/content/10.6...

biorxiv.orgEstimating de novo mutation rates using parent-offspring pairsExisting pedigree approaches to identifying de novo mutations (DNMs) require at least two parents and a single offspring, limiting applicability. Here, we introduce OOPS (Only One Parent Sequencing), ...

Jeff Spence reposted

Jeffrey M Kidd
@jeffreymkidd.bsky.social

This is a very nice review article from @pontus-skoglund.bsky.social and @mathiesoniain.bsky.social . Figure 2 gives a very nice summary of the different ways selection tests are calibrated. They also offer 5 succinct summaries of the current state of the field for humans.

Nature Reviews Genetics@natrevgenet.nature.com

New online! Genomic insights into natural selection in recent human history

dlvr.itGenomic insights into natural selection in recent human historyNature Reviews Genetics, Published online: 11 September 2026; doi:10.1038/s41576-026-01010-9Advances in genome-wide selection scans and ancient DNA have greatly expanded our ability to study human evolution. This Review discusses the conceptual and statistical challenges in identifying and interpreting selection signals, and how recent approaches are reshaping our understanding of human evolution over the past 10,000 years.

Jeff Spence reposted

Yun Deng
@yundeng.bsky.social

I am excited to share that I will be joining the Baker Institute for Animal Health and College of Veterinary Medicine at Cornell University as an Assistant Professor next spring! I am grateful to my mentors and colleagues who have supported me throughout this journey.