Lightnews — Scholar-powered news

Reposted by Brian Hie

Kevin K. Yang 楊凱筌 @kevinkaichuang.bsky.social · 13d

Combine multimer structure prediction and an antibody language model to design de novo antibodies with nanomolar binding affinity.

@synbiogaolab.bsky.social @brianhie.bsky.social

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

3 19

Reposted by Brian Hie

Xiaojing Gao @synbiogaolab.bsky.social · 14d

Having often dealt with the frustration of binder-limited projects, we sought a more accessible source for nanobodies than yeast display or llama. Here we introduce Germinal, computationally designing antibody-like binders with such a hit rate that only tens need to be screened for each target.

1 3 7

Reposted by Brian Hie

Arc Institute @arcinstitute.org · 14d

In another preprint from the @brianhie.bsky.social Lab and @synbiogaolab.bsky.social, they introduce Germinal, a generative AI system for de novo antibody design.

Germinal produces functional nanobodies in just dozens of tests, making custom antibody design more accessible than ever before.

1 14 16

Reposted by Brian Hie

Adrian Woolfson @adrianwoolfson.bsky.social · 21d

A landmark paper from Brian Hie’s group at the Arc Institute. The de novo design of the synthetic genome of an entirely novel biological entity

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

Generative design of novel bacteriophages with genome language models

Many important biological functions arise not from single genes, but from complex interactions encoded by entire genomes. Genome language models have emerged as a promising strategy for designing biol...

www.biorxiv.org

4 11

Reposted by Brian Hie

Kevin K. Yang 楊凱筌 @kevinkaichuang.bsky.social · 21d

Genome language models can generate new, high-fitness bacteriophages!

@samuelhking.bsky.social @claudiadriscoll.bsky.social
@david-li.bsky.social @danguo.bsky.social @adititm.bsky.social Garyk Brixi @maxewilkinson.bsky.social @brianhie.bsky.social

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

2 9 31

Reposted by Brian Hie

Paul Hoskisson 🧫 🦠🐸 @paulhoskisson.bsky.social · 21d

This is super cool from @brianhie.bsky.social et al-
Generative design of novel bacteriophages with genome language models www.biorxiv.org/content/10.1...

Generative design of novel bacteriophages with genome language models

Many important biological functions arise not from single genes, but from complex interactions encoded by entire genomes. Genome language models have emerged as a promising strategy for designing biol...

www.biorxiv.org

3 9

Reposted by Brian Hie

Arc Institute @arcinstitute.org · 21d

In a new preprint from @brianhie.bsky.social's lab, the team reports the first generative design of viable bacteriophage genomes.

Leveraging Evo 1 & Evo 2, they generated whole genome sequences, resulting in 16 viable phages with distinct genomic architectures.

4 12 38

Reposted by Brian Hie

Samuel King @samuelhking.bsky.social · 21d

Many of the most complex and useful functions in biology emerge at the scale of whole genomes.

Today, we share our preprint “Generative design of novel bacteriophages with genome language models”, where we validate the first, functional AI-generated genomes 🧵

3 20 49

Reposted by Brian Hie

The National Academies of Sciences, Engineering and Medicine @nationalacademies.org · Mar 14

#AI is transforming #biology, enabling rapid innovation — but with progress comes risk.

A new National Academies report examines how AI-driven biological tools could impact #biosecurity, from potential misuse to their role in risk mitigation.

Read: buff.ly/2unawE4

2 4

Reposted by Brian Hie

Hannes Stark @hannes-stark.bsky.social · Mar 13

Video is up!
"Genome modeling and design across all domains of life with Evo 2" youtu.be/Rarn97Wpl1A
With the author Garyk Brixi!

Genome modeling and design across all domains of life with Evo 2 | Garyk Brixi

YouTube video by Valence Labs

youtu.be

3 18

Reposted by Brian Hie

Kevin K. Yang 楊凱筌 @kevinkaichuang.bsky.social · Feb 27

Evo2: a fully open genome language model that can predict the functional effects of variation and generate large genomic segments

@garykbrixi.bsky.social @matthewdurrant.com @michaelpoli.bsky.social @genophoria.bsky.social @pdhsu.bsky.social @brianhie.bsky.social

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

1 7 31

Reposted by Brian Hie

Kevin K. Yang 楊凱筌 @kevinkaichuang.bsky.social · Feb 21

All-atom fixed backbone protein sequence design with FAMPNN

@richardshuai.bsky.social Talal Widatalla @possuhuanglab.bsky.social @brianhie.bsky.social

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

7 30

Brian Hie @brianhie.bsky.social · Feb 19

Paper: arcinstitute.org/manuscripts/...

Blog: arcinstitute.org/news/blog/evo2

GitHub: github.com/ArcInstitute...

5

Brian Hie @brianhie.bsky.social · Feb 19

We trained a genomic language model on all observed evolution, which we are calling Evo 2.

The model achieves an unprecedented breadth in capabilities, enabling prediction and design tasks from molecular to genome scale and across all three domains of life.

1 14 29

Reposted by Brian Hie

Quanta Magazine @quantamagazine.bsky.social · Feb 5

Brian Hie harnessed the powerful parallels between DNA and human language to create an AI tool that interprets genomes. Read his conversation with Ingrid Wickelgren: www.quantamagazine.org/the-poetry-f...

The Poetry Fan Who Taught an LLM to Read and Write DNA | Quanta Magazine

By treating DNA as a language, Brian Hie’s “ChatGPT for genomes” could pick up patterns that humans can’t see, accelerating biological design.

www.quantamagazine.org

1 14 40

Brian Hie @brianhie.bsky.social · Jan 1

Thanks Eric!

3

Reposted by Brian Hie

Eric Topol @erictopol.bsky.social · Jan 1

The 2024 Good Tech Awards, spotlighted by @kevinroose.com for #AI/Science
nytimes.com/2024/12/30/t...
congrats!
@arcinstitute.org @brianhie.bsky.social @patrickhsuu.bsky.social @jameszou.bsky.social and the others for well deserved recognition

2 11 64

Reposted by Brian Hie

Sarafan ChEM-H @stanford-chemh.bsky.social · Dec 24

#YearInReview An AI-based method to develop better antibody medicines faster: https://buff.ly/3P74SCo
@brianhie.bsky.social

Institute Scholar Peter Kim, Brian Hie, Varun Shanker, and team’s new AI approach optimizes development of antibody drugs.

1 1

Reposted by Brian Hie

Di Jiang @dijiang319.bsky.social · Dec 22

New @biorxiv-synthbio.bsky.social on #Evo 👀⤵️🧵 introducing Evo 1.5 for semantic mining + SynGenome - an AI-generated genomics database #AI #synbio #LLM🧬 @adititm.bsky.social @brianhie.bsky.social et al. @arcinstitute.org

3 3 11

Brian Hie @brianhie.bsky.social · Dec 20

Yes they appear to be having server problems, here is a link to the paper PDF: evodesign.org/Semantic_Min...

evodesign.org

1 1

Reposted by Brian Hie

Aditi Merchant @adititm.bsky.social · Dec 19

Excited to have the first project of my PhD out!! By leveraging genomic language model Evo’s ability to learn relationships across genes (i.e., "know a gene by the company it keeps"), we show that we can use prompt-engineering to generate highly divergent proteins with retained functionality. 🧵1/N

1 5 19

Reposted by Brian Hie

Kevin K. Yang 楊凱筌 @kevinkaichuang.bsky.social · Dec 18

A protein's position in the genome in relation to other genes is informative about function, so a genomic language model can be prompted to generate
- toxin/antitoxin pairs
- anti-CRISPR proteins

+ a database of 120B synthetic base pairs

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

@brianhie.bsky.social

In-context genomic modeling and design with Evo.

120 billion base pairs of AI-generated genomic sequences with SynGenome.

Evo generates functional anti-CRISPR proteins with no homology to known proteins.

Evo generates functional toxin-antitoxin protein-protein interactions with remote homology to nature.

1 19 60

Brian Hie @brianhie.bsky.social · Dec 17

We apologize for any inconvenience caused. This issue does not affect any of the results of our paper and only affects the evo-model API. Users interfacing via the Hugging Face AutoModelForCausalLM API should be unaffected. Special thanks to @danielchang.bsky.social for pointing us to the problem.

Brian Hie @brianhie.bsky.social · Dec 17

We have found and fixed a bug in the code for Evo model inference affecting package versions from Nov 15-Dec 16, 2024, which has been corrected in the latest release. If you installed the package during this timeframe, please upgrade to correct the issue.

1 4

Brian Hie @brianhie.bsky.social · Nov 16

Thanks Tim!

2