Lightnews — Scholar-powered news

Elizabeth Wood, PhD @lizbwood.bsky.social · 2d

This is an interesting case because by all rights, Sasha Rudensky could have easily been named and won as well; I would be curious as to what Brunkow would say if offered (theoretically) to trade the award for access to a career like Rudensky's, and vice-versa!

1 1

Elizabeth Wood, PhD @lizbwood.bsky.social · 2d

All scientists! But yes, women scientists, too. Ramsdell, who was outside of academia like Brunkow, got to serve as the CSO of Sonoma; Brunkow was a science writer for a while and then a program manager.

2 1

Elizabeth Wood, PhD @lizbwood.bsky.social · 3d

Could it be that it’s so close to translation? There’s an overwhelming feeling of responsibility to stay in, working on what’s important (curing disease, etc.)

1

Elizabeth Wood, PhD @lizbwood.bsky.social · 4d

I’ll delete this slightly snide comment!

1

Elizabeth Wood, PhD @lizbwood.bsky.social · 4d

I’d find it more heartening if they hadn’t *wanted* to stay in academia: Kariko famously so, but Bruknow too. Lee managed it, but as a senior scientist in her husband and coauthor’s lab.

1

Elizabeth Wood, PhD @lizbwood.bsky.social · 4d

I know that these sorts of stories, where recognition comes in the end, are supposed to provide a certain kind of told-you-so satisfaction and inspiration; but I'm sure I (and most) would trade it for 20-40 years of productive work life of controlling and shaping a research program.

14

Elizabeth Wood, PhD @lizbwood.bsky.social · 4d

For every Nobel that goes to a criminally under-recognized woman scientist (Brunkow, Karikó), or fails to go (Candy Lee), a week of mourning and reform for an academic system wherein you can do Nobel-prize-worthy-work and still end up without a conceivable path to being a professor.

Nature @nature.com · 4d

BREAKING: The Nobel Prize in Physiology or Medicine has been awarded jointly to Mary E. Brunkow, Fred Ramsdell, and Shimon Sakaguchi "for their discoveries concerning peripheral immune tolerance"

Stay tuned for more.
#NobelPrize

2 15 44

Elizabeth Wood, PhD @lizbwood.bsky.social · 14d

I was thinking this too! If anyone on here has a warm intro for me... I reached out today to a few people there, but just on LinkedIn.

1 1

Elizabeth Wood, PhD @lizbwood.bsky.social · 14d

Let's say all of a sudden you found yourself with 100,000+ wetlab validated, de novo AI designed, fully human GLP1R binding antibodies? Could you turn the patent suite into a financial instrument? Anyone who thinks about this you could connect me to?

1 1

Reposted by Elizabeth Wood, PhD

JURA Bio, Inc. @jura.bsky.social · 18d

How do you train AI on protein data when functional sequences are astonishingly rare? Traditional iterative optimization fails because most measurements show zero.

Our newly announcing learning system, LeaVS, shows a way.

Elizabeth Wood, PhD @lizbwood.bsky.social · 18d

You can read more in our post at www.jurabio.com/blog/leavs; preprint forthcoming.

@jura.bsky.social @eliweinstein.bsky.social @mgollub.bsky.social @highvariance.bsky.social

LeaVS: Accelerating learning for biological AI — JURA Bio, Inc.

A fundamental lesson of modern AI is that scale is essential: training bigger models on bigger datasets unlocks new capabilities. A fundamental lesson of AI engineering is that scaling up isn't trivia...

www.jurabio.com

1 3

Elizabeth Wood, PhD @lizbwood.bsky.social · 18d

You can read more in our post at www.jurabio.com/blog/leavs; preprint forthcoming.

@jura.bsky.social @eliweinstein.bsky.social @mgollub.bsky.social @highvariance.bsky.social

LeaVS: Accelerating learning for biological AI — JURA Bio, Inc.

A fundamental lesson of modern AI is that scale is essential: training bigger models on bigger datasets unlocks new capabilities. A fundamental lesson of AI engineering is that scaling up isn't trivia...

www.jurabio.com

5 4

Elizabeth Wood, PhD @lizbwood.bsky.social · 18d

Today we introduce LeaVS (Learning from Variational Synthesis), a system for accelerating AI training on this functional data. LeaVS goes beyond conventional model training procedures by exploiting knowledge of the underlying synthetic library.

The result? A dramatic speedup in learning.

1 1

Elizabeth Wood, PhD @lizbwood.bsky.social · 18d

Instead, it requires changes to how we design algorithms, allocate resources, and engineer systems. It requires carefully identifying and removing bottlenecks on learning, across the entire training stack.

1

Elizabeth Wood, PhD @lizbwood.bsky.social · 18d

A fundamental lesson of modern AI is that scale is essential: training bigger models on bigger datasets unlocks new capabilities. A fundamental lesson of AI engineering is that scaling up isn't trivial: it is not just a matter of spending more money and resources.

1 3 7

Reposted by Elizabeth Wood, PhD

henrietteautzen.bsky.social @henrietteautzen.bsky.social · Aug 3

🚨 One week left to apply!
I am hiring a fully funded PhD fellow and a postdoc to work on membrane protein structural biology and pharmacology at Univ Copenhagen

Come work with a great team and exciting projects in a collaborative environment!

🔗 Links to the ads in comments

#CryoEM #AcademicJobs

2 12 19

Reposted by Elizabeth Wood, PhD

Eli Weinstein @eliweinstein.bsky.social · Jul 24

I'm looking for my first PhD student! We will push the frontiers of probabilistic machine learning for the molecular sciences, and study how to design new algorithms that exploit the unique properties of molecular systems to learn about the world.

efzu.fa.em2.oraclecloud.com/hcmUI/Candid...

PhD scholarship in Machine Learning for Molecules - DTU Chemistry

Advance large scale data generation for chemical and biological AI in a 3 year PhD. Work on the frontier of active learning, and develop novel probabilistic machine learning techniques for experiment ...

efzu.fa.em2.oraclecloud.com

7 15

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 24

Want to work at the frontier of active learning and AI-guided experimentation? @eliweinstein.bsky.social is hiring PhD students at @dtu.dk DTU Chemistry.

Full job link: efzu.fa.em2.oraclecloud.com/hcmUI/Candid...

Thanks for re-sharing widely!

PhD scholarship in Machine Learning for Molecules - DTU Chemistry

Advance large scale data generation for chemical and biological AI in a 3 year PhD. Work on the frontier of active learning, and develop novel probabilistic machine learning techniques for experiment ...

efzu.fa.em2.oraclecloud.com

5 6

Reposted by Elizabeth Wood, PhD

Chris Thorpe @drchristhorpe.bsky.social · Jul 23

Totally agree that standard biological data isn't built for AI. We need to think strategically about how we build better, less biased richer datasets specifically for machine learning

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 22

The biggest challenge for AI in biology isn't just models, it's the data used to train them. Standard biological data isn't built for AI. To unlock generative AI for drug discovery, we must rethink how we generate and capture data. 1/

Hardware/wetware codesigned data loop VISTA makes use of generative model sampling and synthesis "on chip" on-board by leveraging oligosynthesis setup shown here.

1 4

Reposted by Elizabeth Wood, PhD

Mattia Gollub @mgollub.bsky.social · Jul 23

Thrilled to demonstrate VISTA, our platform for creating high quality, AI-ready biological data loops at massive scale!

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 22

The biggest challenge for AI in biology isn't just models, it's the data used to train them. Standard biological data isn't built for AI. To unlock generative AI for drug discovery, we must rethink how we generate and capture data. 1/

2 6

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 22

This is what's so transformative about variational-synthesis driven discovery; we don't have to learn from positive results only as we have the underlying generative model sequences in addition to the positive hits.

1

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 22

Great coverage of VISTA by the wonderful @faylinphd.bsky.social over at GEN: www.genengnews.com/topics/artif...

Read about VISTA: www.jurabio.com/blog/vista

Jura Bio’s AI Data Loop Enables Large-Scale De Novo Antibody Design

VISTA, an AI-controlled data loop system, guides large-scale antibody design from scratch for applications in cancer therapy.

www.genengnews.com

3 8

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 22

When AI drives your data generation, learning is more efficient and effective. Take a deep dive into VISTA:

6

Reposted by Elizabeth Wood, PhD

Eli Weinstein @eliweinstein.bsky.social · Jul 22

So excited about this - we did iterative generative design at large scale with variational synthesis, and got human scFv candidates against some of the hardest therapeutic targets around.

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 22

The biggest challenge for AI in biology isn't just models, it's the data used to train them. Standard biological data isn't built for AI. To unlock generative AI for drug discovery, we must rethink how we generate and capture data. 1/

3 7

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 22

You can read the full technical note at: jurabio.com/blog/vista

We're on Bluesky at @jura.bsky.social. My thanks for reposting widely for reach!

8/8

VISTA: high quality, AI-ready functional biological data loops at massive scale — JURA Bio, Inc.

AI holds tremendous potential for advancing drug design and human therapeutics. Generative models provide tools to design and sculpt proteins and other molecules. But existing systems are bottlenecked...

jurabio.com

3 5

Elizabeth Wood, PhD @lizbwood.bsky.social · Jul 22

We're pleased to announce VISTA as a technical note today.

🚨Developed 1,497 diverse, human antibody-based scFv-CARs against 7 key intracellular oncology targets.

🚨Successfully designed highly specific candidates for 3 of these targets that were previously undrugged by TCR-like antibodies.

7/

1 4