William DeWitt

@wsdewitt.github.io

Assistant Professor, University of Washington, Genome Sciences. Previous: JSMF Fellow, Berkeley EECS ♡: Computational biology, evolutionary dynamics, quantitative immunology https://dewitt-lab.github.io/ [disclaimer: opinions mine] 🇺🇸🚫👑

What if an embryo made immune cells before it finished building the tissue they usually come from? In our new preprint, we show that annual killifish embryos specify neutrophils before germ layers are fully resolved. www.biorxiv.org/content/10.6... Huge thanks to my PI, @ohnolog.bsky.social ! 🧵

Early immune cell development precedes gastrulation in annual killifish

During embryogenesis, cell types arise in a predictable order because developmental regulators act sequentially. But how evolutionary changes in morphogenesis reshape the signaling environments that a...

biorxiv.org

The human genome is no blueprint or algorithm — it’s something else entirely, something living with promiscuous logic. Can new genomic AI models, such as Evo 2 and AlphaGenome, untangle these tricky knots or will the genome confound them, too? www.quantamagazine.org/why-the-huma...

Why the Human Genome’s Tangled Physicality May Confound AI | Quanta Magazine

Our genetic heritage is not a blueprint or an algorithm, as many biologists have imagined, but something else entirely.

quantamagazine.org

I've learned a lot from this community at various meetings since I was a PhD student. I'm looking forward to this one!

John Tsang@tsanglab.bsky.social · 2mo ago

6th iteration of the Physics & Computation in Immunology meeting will be at Yale CSEI @yalecsei.bsky.social Sep 28–30. With Arup Chakraborty, Michael Lässig, @thearmita.bsky.social & Alex Walczak. Speakers from Princeton, Cambridge, Harvard, MSK, UCSF & more. Free. 120 spots. Deadline Sep 6.

Antibody LMs learn what looks antibody-like, but not how selection turns naive germline antibodies into strong binders. @aakarshv1.bsky.social and I are excited to share CoSiNE, a model that learns this germline-to-mature process for variant effect prediction and antibody design. (1/8)

This was one of the most complex projects I have ever been involved in. Took us the better part of a decade, but I thoroughly enjoyed every minute of it. Especially the intensive and extensive interactions with @wsdewitt.github.io, @matsen.bsky.social, and @tylernstarr.bsky.social

William DeWitt@wsdewitt.github.io · 2mo ago

This started in 2019 as daydreamy PhD student musings with Tatsuya Araki. We're as excited as ever about germinal centers as a platform for experimental evolution. Many thanks to PIs @victora.bsky.social @matsen.bsky.social, co-1st authors Ashni Vora and Tatsuya, and many other key collaborators!

This started in 2019 as daydreamy PhD student musings with Tatsuya Araki. We're as excited as ever about germinal centers as a platform for experimental evolution. Many thanks to PIs @victora.bsky.social @matsen.bsky.social, co-1st authors Ashni Vora and Tatsuya, and many other key collaborators!

Replaying germinal center evolution on a quantified affinity landscape

Antibody affinity maturation results from a somatic evolutionary process that takes place in the germinal center. A “parallel replay” experiment on germinal center B cells reveals the evolutionary for...

cell.com

Out now in Science! Our study challenges long-standing assumptions about transcription factor specificity in eukaryotes. Novel single-molecule measurements of TF behavior in living cells reveal an independence of locus-specific binding from DNA sequence recognition.🧵 www.science.org/doi/10.1126/...

Unstructured transcription factor interactions enable emergent specificity

How intrinsically disordered regions (IDRs) shape chromatin binding and nuclear organization of transcription factors (TFs) remains unclear. We used proximity-assisted photoactivation (PAPA), a single...

science.org