Sign in

markgerstein.bsky.social

@markgerstein.bsky.social
111 followers 100 following 37 posts
PostsRepliesMedia
Reposted by @markgerstein.bsky.social
Yale MB&B @yalembb.bsky.social · 23/09/2026
We are excited to announce a tenure-track position in Computational Biophysics and Biochemistry, exploring the intricate molecular and cellular processes and the complex interactions of their macromolecular components! apply.interfolio.com/193444
apply.interfolio.com
Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio
027
markgerstein.bsky.social @markgerstein.bsky.social · 23/09/2026
Posting a short talk tomorrow for the Quantum World Congress on using Quantum Computing to achieve Biomedical Privacy lectures.gersteinlab.org/summary/Quan... #QWC2026
000
markgerstein.bsky.social @markgerstein.bsky.social · 29/07/2026
Posting my talk tomorrow at Oxford CS on AI in Structural Bioinformatics lectures.gersteinlab.org/summary/AI-S... Modelling Flexibility, Disorder & Interactions. Lots of new slides on World Models for protein folding.
020
markgerstein.bsky.social @markgerstein.bsky.social · 24/07/2026
Posting my talk today at the University of Montpellier on the changing challenges in genomics lectures.gersteinlab.org/summary/Chan... Lots of slides on variant impact models
000
markgerstein.bsky.social @markgerstein.bsky.social · 20/07/2026
Posting my talk tomorrow at @crg.eu in Barcelona (hosted by @beaborsari.bsky.social & @guigolab.bsky.social) lectures.gersteinlab.org/summary/Usin... Lots of slides on the AI Scientist
000
markgerstein.bsky.social @markgerstein.bsky.social · 20/07/2026
Posting my talk today on the 25th anniversary of the human genome (hosted by @beaborsari.bsky.social) lectures.gersteinlab.org/summary/Chan... Lots of new slides on variant impact models & pseudogene epigenetics across tissues
010
markgerstein.bsky.social @markgerstein.bsky.social · 12/07/2026
Posting my talk today for tutorial IP2 at @iscb.bsky.social's #ISMB2026: LLMs & Agentic AI for Biomedical Informatics (organized by R Tang) lectures.gersteinlab.org/summary/LLMs... Has lots of slides on case studies for LLMs in brain genomics & comparisons with more classic approaches
042
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 23/06/2026
During our recent lab roster meeting, we asked about everyone's AI chatbot preferences (n=69, multi-select): 🥇 Claude — 49 (71%) 🥈 ChatGPT — 43 (62%) 🥉 Gemini — 14 (20%) Takeaway: Mostly Claude + ChatGPT co-usage, with Gemini as a distant third and a long tail of niche tools.
111
markgerstein.bsky.social @markgerstein.bsky.social · 18/06/2026
Post term, cleaned up my biomed. data sci. course website. gersteinlab.org/courses/3520 Now in its 28th year. New thing this year was #AI integration: the students merged (with color coding) the manual & AI summaries of each lecture. Also, lots of new posted videos & PDFs (esp. on DL)
100
markgerstein.bsky.social @markgerstein.bsky.social · 04/06/2026
Posting my talk tomorrow at #FOGBoston (#FOG26) lectures.gersteinlab.org/summary/Mult... Lots of slides on single-cell multi-omics integration for brain disorders. New slides on a formalism for model interpretation @flgenomics.bsky.social @fogenomics.bsky.social
010
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 01/06/2026
New @natcomms.nature.com paper by @katebowie.bsky.social @markgerstein.bsky.social, where we study how disinfection shapes microbes in hospital sink drain biofilms. Biofilms regrew in 4 days, enriched for carbapenem-resistant bacteria and multidrug efflux pump genes: www.nature.com/articles/s41...
143
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 01/06/2026
In our @natmachintell.nature.com paper, we introduce a framework to analyse interpretability in deep learning by drawing on a formal notion of model semantics from the philosophy of science. We illustrate our framework with examples from biomedicine. Read here: rdcu.be/e9uYh
211
markgerstein.bsky.social @markgerstein.bsky.social · 20/05/2026
Thoughts on Robert Wachter's Giant Leap blog.gerstein.info/2026/05/thou... Great points about AI's early missteps vs current successes in medicine
000
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 06/05/2026
Curious how pseudogenes are transcriptionally regulated? Our new Genome Research paper shows processed pseudogenes break the rules: they’re transcribed without classic epigenetic marks, linked to enhancers, and enriched for YY1 motifs. Study co-led by Yunzhe Jiang and @beaborsari.bsky.social
doi.org
Epigenetic characterization of pseudogenes across human tissues
Pseudogenes have historically been regarded as nonfunctional remnants of genome evolution. However, relative to other noncoding genomic elements, their promoter architecture and epigenetic regulation remain incompletely understood. Here, we systematically characterize pseudogene promoters and compare them with those of protein-coding genes and long noncoding RNAs. To do this, we integrate matched transcriptomic and epigenomic data across 26 human tissues from the EN-TEx (ENCODE-GTEx) project. We uniformly annotate promoters with chromatin features (histone modifications, chromatin accessibility, and DNA methylation), sequence motifs, and evolutionary conservation, generating an online catalog. Leveraging this catalog, we show that, across multiple tissues, transcribed, unprocessed pseudogenes exhibit chromatin patterns similar to those of active protein-coding genes. In contrast, transcribed, processed pseudogenes show a different pattern: most lack the canonical hallmarks of transcription (e.g., active histone marks) at their promoters. Instead, their promoters show increased overlap with LINE elements, enrichment for YY1-like binding motifs, and higher Hi-C contact frequency, particularly with distal enhancer-like regulatory regions. Together with their greater conservation (relative to unprocessed pseudogenes), these features suggest that the transcription of processed pseudogenes may require regulatory mechanisms distinct from canonical promoter-associated epigenetic activation.
022
markgerstein.bsky.social @markgerstein.bsky.social · 16/03/2026
Posting my talk tomorrow at St. Johns U. on brain genomics lectures.gersteinlab.org/summary/Clas... Lots of new slides on the challenges of AI model applicability to genomics & on private federated learning with co-variate matching.
021
markgerstein.bsky.social @markgerstein.bsky.social · 08/03/2026
Posting my talk tomorrow @nygenome.org (hosted by @gamzeandgursoy.bsky.social). lectures.gersteinlab.org/summary/AI-A... Doing genomics with AI & then seeing if AI can take over the whole thing (spoiler: not yet!). New slides on assessing agent models with externalized reasoning (vs. baselines)
062
Reposted by @markgerstein.bsky.social
Techmeme @techmeme.com · 28/01/2026
Google DeepMind researchers unveil AlphaGenome, an AI model trained on molecular data to predict 11 different genomic processes, such as gene splicing (Carl Zimmer/New York Times) Main Link | Techmeme Permalink
052
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 03/12/2025
Our @neuripsconf.bsky.social work led by YunyangLI “E2Former: An Efficient and Equivariant Transformer with Linear-Scaling Tensor Products” was selected as a spotlight (with score ranked ~17 / 21k submissions). Poster: Thur Dec 4, Exhibit Hall CDE #5512 Online: openreview.net/forum?id=ls5...
111
markgerstein.bsky.social @markgerstein.bsky.social · 29/12/2025
Thoughts on @jsellenberg.bsky.social's Shape blog.gerstein.info/2025/12/thou... Engaging Stories about Hard Stuff, esp. liked the Discussion of Math behind AI
000
markgerstein.bsky.social @markgerstein.bsky.social · 02/12/2025
Posting my talk tomorrow @umich.edu, on analyzing endophenotypes (expression, biosensors & imaging). lectures.gersteinlab.org/summary/Geno... Lots of new slides on using the @gtexportal.bsky.social images to predict age
010
Reposted by @markgerstein.bsky.social
Yale MB&B @yalembb.bsky.social · 26/11/2025
We are excited to announce a tenure-track or tenured faculty position in Computational Biophysics and Biochemistry, exploring the intricate molecular and cellular processes and the complex interactions of their macromolecular components! apply.interfolio.com/177120
apply.interfolio.com
Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio
077
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 20/11/2025
Our new PNAS study bridges histology and genomics. Using deep learning and imageQTL analysis, we show how tissue images reflect gene expression and aging — making histology more interpretable with AI. By RanMeng, W. Zhu, C. Cameron, P. Ni, X. Zhou, T. Ulammandakh, and @markgerstein.bsky.social
pnas.org
PNAS
Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...
041
markgerstein.bsky.social @markgerstein.bsky.social · 20/11/2025
Posting my talk at UW Madison (the DeMets lecture, hosted by @daifengwang.bsky.social). Lots on bioinformatics methods for studying the brain: classic, present (DL) & future (quantum) lectures.gersteinlab.org/summary/Clas... New slides on AI agents & homomorphic encryption for privacy
010
Reposted by @markgerstein.bsky.social
Daniel A. Olivas @danielolivas.bsky.social · 16/11/2025
Can AI developers avoid Frankenstein’s fateful mistake? www.latimes.com/opinion/stor... #AI #Frankenstein
latimes.com
Contributor: Can AI developers avoid Frankenstein’s fateful mistake?
As Guillermo del Toro's adaptation reaches millions this month, its lesson remains urgent: Don't abandon the dangerous things you create.
011
markgerstein.bsky.social @markgerstein.bsky.social · 10/11/2025
Posting my talk tomorrow at the Computational Molecular Sciences & Engineering (CMSE) Symposium @Yale lectures.gersteinlab.org/summary/Enha... Lots of new slides on the discard-and-restart approach for enhanced structural sampling
020
markgerstein.bsky.social @markgerstein.bsky.social · 05/11/2025
Posting my talk today for the AI in Health Series lectures.gersteinlab.org/summary/AI-M... Goes over an evolution of computational methods for biomedicine: Classical, DL & Quantum.
010
markgerstein.bsky.social @markgerstein.bsky.social · 16/10/2025
Posting my talk tomorrow on Biomed AI @mila-quebec.bsky.social lectures.gersteinlab.org/summary/Biom... Some new slides from Yunzhe Jiang on ENTEx & variant impact prediction and W Lee on ensembling ML models for drug screening
000
markgerstein.bsky.social @markgerstein.bsky.social · 09/10/2025
Posting my talk tomorrow on AI in Biomedicine at Einstein Med (hosted by @zhengdy.bsky.social) lectures.gersteinlab.org/summary/AI-t... Using AI for brain genomics & thinking about how it can do this research autonomously. Lots of new slides on AI coding & risks
000
Reposted by @markgerstein.bsky.social
DrDinD.bsky.social @drdind.bsky.social · 07/10/2025
🧵The Nobel Prize in Physics today went to John Clarke (UK), Michel Devoret (France), & John Martinis (USA) today for work conducted in Clarke's UC Berkeley lab in the mid 80s showing that alternation of a quantum state could proceed from one side to the other of a device you can see with your eye.
John Clarke, UC Berkeley Emeritus professor, Michel Devoret, Yale Emeritus Professor, and John Martinis, UC Santa Barbara Emeritus Professor, winners of the 2025 Nobel Prize in Physics.
29929
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 22/09/2025
📚 Yale students have returned to campus, so time for a roster meeting! We again made our Nobel Prize predictions (given how accurate we were last year 😉) 🥇Our top prediction is Habener & Knudsen (GLP-1) with 28.5% of the vote! 🥈 In second is Rothberg & David Klenerman (NGS)
111
markgerstein.bsky.social @markgerstein.bsky.social · 06/09/2025
Did my annual triathlon today in Madison. Good swim. Steady bike. Walk-skip-walk to the finish line. Made it! And, unlike last year, did an extensive stretch afterwards & drove myself home. Before & after race pics. (More at flickr.com/photos/mbgmb...)
020
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 25/08/2025
Curious how your favorite gene changes when and how during a biological process? Want to dive into the kinetics of chromatin + gene expression? Meet chronODE, our new tool to model multi-omic time-series with logistic equations + ML! doi.org/10.1038/s414...
doi.org
The chronODE framework for modelling multi-omic time series with ordinary differential equations and machine learning - Nature Communications
Here, the authors use a simple equation to study how genes and their regulators switch on/off over time, across the whole genome in tissues and cells. Most changes are gradual, but some genes switch q...
122
markgerstein.bsky.social @markgerstein.bsky.social · 18/08/2025
Posting my talk on AI & Brain Disorders at Princeton CHDI on Tuesday lectures.gersteinlab.org/summary/Ext-... Lots of new slides on drug repurposing and predicting age from histology images
000
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 15/07/2025
🧠 At our recent Gerstein Lab roster meeting, we took a detour into… personality science! Turns out we’re INT Central 🧪 📌 70% Introverts 📌 83% Intuitives 📌 57% Thinkers Analysts (INTP, INTJ) dominate, far more than the U.S. baseline. #MBTI #INTP #INTJ
111
markgerstein.bsky.social @markgerstein.bsky.social · 03/08/2025
Thoughts on Pearl's Book of Why blog.gerstein.info/2025/08/thou... Lots of Good Intuition on Bayesian Networks, Sometimes with Strong Opinions
000
Reposted by @markgerstein.bsky.social
Eric Topol @erictopol.bsky.social · 30/07/2025
"The microplastic (MP) 1–10 μm exposure estimates are 100-fold higher than previous estimates that were extrapolated from larger MP sizes, and suggest that the health impacts of MP inhalation may be more substantial than we realize." journals.plos.org/plosone/arti...
journals.plos.org
Human exposure to PM10 microplastics in indoor air
The ubiquitous presence of airborne microplastics (MPs) in different indoor environments prompts serious concerns about the degree to which we inhale these particles and their potential impact on huma...
611940
markgerstein.bsky.social @markgerstein.bsky.social · 30/07/2025
Posting my talk tomorrow at @emblebi.bsky.social‬ (for the @ensembl.bsky.social‬ group) lectures.gersteinlab.org/summary/Expl... AI Approaches for Understanding Brain Disorders: Exploring Intermediate Endophenotypes connecting Genotype to Diseases
020
markgerstein.bsky.social @markgerstein.bsky.social · 28/07/2025
Posting my talk tomorrow at @geneticscam.bsky.social on AI for Brain Disorders lectures.gersteinlab.org/summary/Mode... Lots of new slides on characterizing brain developmental time series with sigmoids
020
markgerstein.bsky.social @markgerstein.bsky.social · 19/07/2025
Posting my talk on Mon. on Cancer Genomics at the Institut Curie lectures.gersteinlab.org/summary/Char... Lots of new slides on using LLMs for "tumor boards" & for generating genomics workflows
000
markgerstein.bsky.social @markgerstein.bsky.social · 26/06/2025
Posting my talk tomorrow at the AI4Chemistry Summit at NYU lectures.gersteinlab.org/summary/AI-f... Lots of new slides on computing small molecule Hamiltonians & predicting macromolecular misfolding
010
markgerstein.bsky.social @markgerstein.bsky.social · 23/06/2025
Posting my talk tomorrow at the SCORCH consortium meeting lectures.gersteinlab.org/summary/Chro... Tools for brain single-cell analysis: ChronODE for series data and ASTRO for spatial data.
000
Reposted by @markgerstein.bsky.social
Yale MB&B @yalembb.bsky.social · 09/06/2025
Join us next week for MB&B's Computational Biophysics & Biochemistry Symposium on 6/17 in Bass 305!
022
markgerstein.bsky.social @markgerstein.bsky.social · 09/06/2025
Posting my talk tomorrow for the IMPACT-MH Consortium lectures.gersteinlab.org/summary/Digi... Digital Phenotyping from Wearables Using AI. Some new slides on using wearables to predict exercise lapses
000
markgerstein.bsky.social @markgerstein.bsky.social · 07/06/2025
School's out for the summer, & I've finally got around to cleaning up the website for my class on biomedical data science (gersteinlab.org/courses/452) I've been teaching this course for >25 years. This year, we've posted lots of new lecture videos.
100
Reposted by @markgerstein.bsky.social
Yale Ventures @yaleventures.bsky.social · 08/05/2025
📣 200+ Innovators Presenting at the 2025 Yale Innovation Summit 📣 Meet the founders who will compete for $300k in prizes during 6 live pitch sessions, interactive ePoster presentations, and virtual showcases at #YaleInnovation2025 👉 bit.ly/3GKJ0Mk
031
markgerstein.bsky.social @markgerstein.bsky.social · 27/05/2025
Liked the visual effect of @newyorker.com's recent "Double Cover"
000
markgerstein.bsky.social @markgerstein.bsky.social · 15/05/2025
Posting my talk tomorrow at the @broadinstitute.org's Stanley Center lectures.gersteinlab.org/summary/Expl... Exploring Intermediate Endophenotypes connecting Genotype to Diseases
010
Reposted by @markgerstein.bsky.social
gersteinlab.bsky.social @gersteinlab.bsky.social · 05/05/2025
1/4 🚀 New #ICLR2025 SPOTLIGHT ALERT Gerstein Lab presents “Enhancing the Scalability & Applicability of Kohn-Sham Hamiltonians”—led by YunyangLI & Z Xia & L Huang & J Zhang & @markgerstein.bsky.social . Joint work with @msftresearch.bsky.social .
121
markgerstein.bsky.social @markgerstein.bsky.social · 06/05/2025
Posting my talk today at the @ucirvine.bsky.social center for Neural Circuit Mapping lectures.gersteinlab.org/summary/SCZ-... . Lots of slides on using AI approaches for Schizophrenia, Alzheimer's & ADHD, including some new ones on analyzing AD imaging.
000
markgerstein.bsky.social @markgerstein.bsky.social · 04/05/2025
Posting my talk on Mon. for the Frontiers in Computational Biosciences series at @UCLA lectures.gersteinlab.org/summary/Bett... . Lots of new slides on creating digital phenotypes from wearable devices & relating them to genetics and on quantum computing
000