正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

We recently released a technical report highlighting how Isomorphic Labs’ drug design engine is achieving a step change in how we model complex biological mechanisms, consistently outperforming industry benchmarks. Read our blog to discover how we’re bridging the gap between structure prediction and rational drug design. You can find... show more

Isomorphic Labs

65,908 subscribers

125,621 次观看 • 8 个月前 •via X (Twitter)

健康养生教育科学技术

Anya Rossi• Live Now

Private livecam show

12 条评论

Daniel Bar 的头像

Daniel Bar8 个月前

If it's not open source, it's just PR.

DULA 的头像

DULA8 个月前

@IsomorphicLabs please check this research 🔬

CRISPRKing 的头像

CRISPRKing8 个月前

So you can now find cryptic pockets from sequence alone. What percentage of your active programs have changed target site because of this?

nerdy penguin🥰🤗 的头像

nerdy penguin🥰🤗8 个月前

"High-performance antibody-antigen binding modeling" a dream scenario for the pharmaceutical industry. No more #Covid like attack.

R. Dumas 的头像

R. Dumas8 个月前

The gap between structure prediction and rational drug design is where the real challenge lies. Consistently beating benchmarks speaks volumes about genuine progress in computational biology.

Ⓜⓐⓡⓖⓘⓝ ⓒⓐⓛⓛ 的头像

Ⓜⓐⓡⓖⓘⓝ ⓒⓐⓛⓛ8 个月前

Any how many drug molecules you have developed or designed so far? Keep playing movies 😂

⚡Jessy SEO formation #IA générative #chatGPT 🦊 的头像

⚡Jessy SEO formation #IA générative #chatGPT 🦊8 个月前

Great, good job

nerdy penguin🥰🤗 的头像

nerdy penguin🥰🤗8 个月前

much better drug design AI

Prince🚀 的头像

Prince🚀8 个月前

Cool

NewAIWorld 的头像

NewAIWorld8 个月前

Biology is the most complex software in the universe. We are finally moving from being users who are confused by the interface to being the developers who can rewrite the source code.

Abenezer 的头像

Abenezer8 个月前

If it can't cure baldness, It's a dime a dozen.

Dee Scientist 的头像

Dee Scientist8 个月前

This is great! Thanks for sharing.

相关视频

We’ve been developing AlphaFold 3 to advance how we do drug design at Isomorphic Labs. This model allows us to a) do rational structure based drug design in-silico, and b) understand more about the biological context of targets 1/

We’ve been developing AlphaFold 3 to advance how we do drug design at Isomorphic Labs. This model allows us to a) do rational structure based drug design in-silico, and b) understand more about the biological context of targets 1/

Max Jaderberg

25,806 次观看 • 2 年前

Introducing #mBER: our open-source AI library for de novo, format-aware antibody design. Paired with in vivo multiplexed screening, mBER forms Manifold Bio's direct-to-vivo drug discovery platform, bridging the gap between AI-guided design and biological translation.

Introducing #mBER: our open-source AI library for de novo, format-aware antibody design. Paired with in vivo multiplexed screening, mBER forms Manifold Bio's direct-to-vivo drug discovery platform, bridging the gap between AI-guided design and biological translation.

Manifold Bio

15,504 次观看 • 1 年前

$"Across our programs, we've advanced quality novel candidates after synthesizing only around 300 compounds, roughly 90% fewer than the industry average." Now that we have mapped the biology, we need to design the exact molecule to target it. In Part 5 of our platform series, SVP of Drug Design Donovan Chin, Ph.D., and VP of Pharmacology Martin Redhead, Ph.D., break down how we transition from biological target discovery to navigating complex chemical space. Just like our approach to biology, we solve chemistry problems through a continuous feedback loop between computation and experimentation. Before a single molecule is ever synthesized in the lab, our team uses Centaur, our integrated drug design environment. By combining hundreds of machine learning and physics-based models into a single platform, we can iteratively design novel molecular structures starting directly from target properties. This generative de novo design approach automatically updates our models so they improve with every single cycle. Our AI-native product engine has already produced over 10 development candidates. By holding our chemistry to the same rigorous standards as our biology, we can synthesize a fraction of the traditional compound volume while delivering superior novel molecules. We believe this focus on quality and novelty over sheer volume will define the true impact of AI on drug development. 🚀 Follow along over the coming weeks as we continue this series, covering how Recursion maps biology, optimizes precision drug design, and closes the gap between patient and discovery. #TechBio #DrugDesign #AIDrugDiscovery #DrugDiscovery #Chemistry$

"Across our programs, we've advanced quality novel candidates after synthesizing only around 300 compounds, roughly 90% fewer than the industry average." Now that we have mapped the biology, we need to design the exact molecule to target it. In Part 5 of our platform series, SVP of Drug Design Donovan Chin, Ph.D., and VP of Pharmacology Martin Redhead, Ph.D., break down how we transition from biological target discovery to navigating complex chemical space. Just like our approach to biology, we solve chemistry problems through a continuous feedback loop between computation and experimentation. Before a single molecule is ever synthesized in the lab, our team uses Centaur, our integrated drug design environment. By combining hundreds of machine learning and physics-based models into a single platform, we can iteratively design novel molecular structures starting directly from target properties. This generative de novo design approach automatically updates our models so they improve with every single cycle. Our AI-native product engine has already produced over 10 development candidates. By holding our chemistry to the same rigorous standards as our biology, we can synthesize a fraction of the traditional compound volume while delivering superior novel molecules. We believe this focus on quality and novelty over sheer volume will define the true impact of AI on drug development. 🚀 Follow along over the coming weeks as we continue this series, covering how Recursion maps biology, optimizes precision drug design, and closes the gap between patient and discovery. #TechBio #DrugDesign #AIDrugDiscovery #DrugDiscovery #Chemistry

Recursion

1,018,443 次观看 • 1 个月前

AI agents can talk, but can they pay? Discover how XDC AI is bridging the gap between chat and real-world transactions to finally unlock AI monetization. A game-changer for the industry!

AI agents can talk, but can they pay? Discover how XDC AI is bridging the gap between chat and real-world transactions to finally unlock AI monetization. A game-changer for the industry!

XDC Network News

19,334 次观看 • 3 个月前

For our 3rd competition for #NeurIPS2023, we’re asking: can you predict a new algorithm for how cells respond to a drug treatment? Learning how #singlecells respond to perturbations will help support more efficient drug design. See more on Kaggle:

For our 3rd competition for #NeurIPS2023, we’re asking: can you predict a new algorithm for how cells respond to a drug treatment? Learning how #singlecells respond to perturbations will help support more efficient drug design. See more on Kaggle:

Cellarity

25,504 次观看 • 3 年前

Very proud to showcase our latest demo of how our Engineering World Model SKY-001 can design a complex Jaguar car with 1000s of complex parts. In order to get us to Autonomous Manufacturing, we need to design systems that can think like a ‘AI General Engineer’: planning complex physical structures, long term sequences, and feasibility directly in the physical domain rather than approximating it through next-token prediction. Engineering design operates in the spatial reasoning domain through hierarchical decomposition, constraint satisfaction, and the sequential construction of complex physical objects. World models are going to be the next frontier in AI after LLMs and SKY-001 is just the beginning.

Very proud to showcase our latest demo of how our Engineering World Model SKY-001 can design a complex Jaguar car with 1000s of complex parts. In order to get us to Autonomous Manufacturing, we need to design systems that can think like a ‘AI General Engineer’: planning complex physical structures, long term sequences, and feasibility directly in the physical domain rather than approximating it through next-token prediction. Engineering design operates in the spatial reasoning domain through hierarchical decomposition, constraint satisfaction, and the sequential construction of complex physical objects. World models are going to be the next frontier in AI after LLMs and SKY-001 is just the beginning.

Sam Pasupalak

82,237 次观看 • 8 天前

Today Google DeepMind released a state of the art computer use model, in partnership with Browserbase. Computer use is hard to evaluate. You need reliable browser infrastructure and realistic tasks. Below, we cover how we ran these benchmarks and how you can try yourself!

Today Google DeepMind released a state of the art computer use model, in partnership with Browserbase. Computer use is hard to evaluate. You need reliable browser infrastructure and realistic tasks. Below, we cover how we ran these benchmarks and how you can try yourself!

Browserbase

155,224 次观看 • 1 年前

"Our integrated system narrows millions of molecules down to a short list that's synthetically accessible. Not just promising in silico, but actually makeable before a single one is sent to the labs." Traditional drug design often leans on decades of precedent and familiar chemistry. In Part 6 of our platform series, SVP of Drug Design Donovan Chin, Ph.D., explains how Recursion breaks that mold by getting a full computational design cycle up and running in about a day. Our AI-native engine vastly increases the aperture for our chemists. Instead of relying solely on known binding pockets, they can efficiently explore entirely novel and complex chemical space. By leveraging our integrated data, the team can forecast non-obvious design hypotheses and predict how to synthesize molecules even when nothing like them currently exists. This process begins by scoring binding affinity using advanced 3D structural models like Nesso-1. From there, the team runs hundreds of integrated models covering ADME, potency, advanced simulations, and AI retrosynthesis. This unified system rapidly filters millions of potential molecules down to a precise shortlist of high-quality candidates that are physically ready for the lab. 🚀 Follow along over the coming weeks as we continue this series, covering how Recursion maps biology, optimizes precision drug design, and closes the gap between patient and discovery. #TechBio #DrugDesign #ComputationalChemistry #DrugDiscovery #MachineLearning

"Our integrated system narrows millions of molecules down to a short list that's synthetically accessible. Not just promising in silico, but actually makeable before a single one is sent to the labs." Traditional drug design often leans on decades of precedent and familiar chemistry. In Part 6 of our platform series, SVP of Drug Design Donovan Chin, Ph.D., explains how Recursion breaks that mold by getting a full computational design cycle up and running in about a day. Our AI-native engine vastly increases the aperture for our chemists. Instead of relying solely on known binding pockets, they can efficiently explore entirely novel and complex chemical space. By leveraging our integrated data, the team can forecast non-obvious design hypotheses and predict how to synthesize molecules even when nothing like them currently exists. This process begins by scoring binding affinity using advanced 3D structural models like Nesso-1. From there, the team runs hundreds of integrated models covering ADME, potency, advanced simulations, and AI retrosynthesis. This unified system rapidly filters millions of potential molecules down to a precise shortlist of high-quality candidates that are physically ready for the lab. 🚀 Follow along over the coming weeks as we continue this series, covering how Recursion maps biology, optimizes precision drug design, and closes the gap between patient and discovery. #TechBio #DrugDesign #ComputationalChemistry #DrugDiscovery #MachineLearning

Recursion

1,091,744 次观看 • 29 天前

Shared at our flagship Research@ event today: In a new technical report, we demonstrate how combining domain-specific foundation models, spatial tools, and upfront planning enables a Geospatial Reasoning agent to carry out complex spatial analysis and planetary prediction that standalone LLMs can’t. This and more on the blog →

Shared at our flagship Research@ event today: In a new technical report, we demonstrate how combining domain-specific foundation models, spatial tools, and upfront planning enables a Geospatial Reasoning agent to carry out complex spatial analysis and planetary prediction that standalone LLMs can’t. This and more on the blog →

Google Research

20,865 次观看 • 28 天前

New tutorial, in this one I tried to show how you can model or design tricky shapes without overcomplicating the process, the video is in the comments.

New tutorial, in this one I tried to show how you can model or design tricky shapes without overcomplicating the process, the video is in the comments.

Mert İskenderoglu

61,129 次观看 • 1 个月前

"This is only possible because we control the design, make, and test steps within one connected system rather than handing off between disconnected tools and teams." Once molecules are designed in silico, they must be synthesized and tested in the real world. In Part 7 of our platform series, VP of Pharmacology Martin Redhead, Ph.D., explains how these candidate designs move directly into our automated DMTL (Design, Make, Test, Learn) labs. Recursion’s automated chemistry platform goes far beyond simple one-pot reactions. We execute complex, multi-step medicinal chemistry at scale. Advanced automation handles the workup, purification, and digital sample tracking for thousands of compounds, creating a digital record of every single experiment. Our in-house AI agents take this a step further by actively helping scientists plan efficient synthetic routes. Because we control the entire design, make, and test lifecycle within one fully integrated system, we completely eliminate the friction and time lost handing off data between disconnected teams. 🚀 Follow along over the coming weeks as we continue this series, covering how Recursion maps biology, optimizes precision drug design, and closes the gap between patient and discovery. #TechBio #DrugDesign #ComputationalChemistry #DrugDiscovery #MachineLearning

"This is only possible because we control the design, make, and test steps within one connected system rather than handing off between disconnected tools and teams." Once molecules are designed in silico, they must be synthesized and tested in the real world. In Part 7 of our platform series, VP of Pharmacology Martin Redhead, Ph.D., explains how these candidate designs move directly into our automated DMTL (Design, Make, Test, Learn) labs. Recursion’s automated chemistry platform goes far beyond simple one-pot reactions. We execute complex, multi-step medicinal chemistry at scale. Advanced automation handles the workup, purification, and digital sample tracking for thousands of compounds, creating a digital record of every single experiment. Our in-house AI agents take this a step further by actively helping scientists plan efficient synthetic routes. Because we control the entire design, make, and test lifecycle within one fully integrated system, we completely eliminate the friction and time lost handing off data between disconnected teams. 🚀 Follow along over the coming weeks as we continue this series, covering how Recursion maps biology, optimizes precision drug design, and closes the gap between patient and discovery. #TechBio #DrugDesign #ComputationalChemistry #DrugDiscovery #MachineLearning

Recursion

167,691 次观看 • 22 天前

We recently shipped Shopify Design. Here's a technical breakdown of how I built it. 🧵

We recently shipped Shopify Design. Here's a technical breakdown of how I built it. 🧵

Johnny Slack

351,567 次观看 • 6 个月前

Today we're launching Latent-Y: the world's first autonomous agent for drug design, lab-validated end to end. Give it a research goal. Latent-Y reasons, designs, iterates, and delivers lab-ready antibodies, autonomously or collaboratively, with the biological reasoning of a PhD protein design expert. Technical report: Blog post: Apply for access:

Today we're launching Latent-Y: the world's first autonomous agent for drug design, lab-validated end to end. Give it a research goal. Latent-Y reasons, designs, iterates, and delivers lab-ready antibodies, autonomously or collaboratively, with the biological reasoning of a PhD protein design expert. Technical report: Blog post: Apply for access:

Simon Kohl

121,784 次观看 • 7 个月前

"A hypothesis from a model isn't a discovery yet. So, it goes back into the lab. The biology the model predicted is the biology we saw at the bench." A computational prediction alone does not cure a disease. In Part 4 of our platform series, Associate Principal Scientist Celia Fernandez explains how Recursion closes the loop to turn AI generated hypotheses into validated biological targets. Once our foundation models point to a potential target, we run the lab in the opposite direction. The hypothesis enters a validation cascade that becomes progressively more demanding and highly specific to the disease. This physical testing allows us to aggressively derisk our computational discoveries directly at the bench. We test these targets against the exact same high bar that our global pharma partners hold their own internal programs to. When the loop finally closes, we can confidently prove that the biology the machine learning model predicted matches the physical reality we see in the lab. 🚀Follow along over the coming weeks as we continue this series, covering how Recursion maps biology, optimizes precision drug design, and closes the gap between patient and discovery. #TechBio #AIdrugdiscovery #AI #drugdiscovery

"A hypothesis from a model isn't a discovery yet. So, it goes back into the lab. The biology the model predicted is the biology we saw at the bench." A computational prediction alone does not cure a disease. In Part 4 of our platform series, Associate Principal Scientist Celia Fernandez explains how Recursion closes the loop to turn AI generated hypotheses into validated biological targets. Once our foundation models point to a potential target, we run the lab in the opposite direction. The hypothesis enters a validation cascade that becomes progressively more demanding and highly specific to the disease. This physical testing allows us to aggressively derisk our computational discoveries directly at the bench. We test these targets against the exact same high bar that our global pharma partners hold their own internal programs to. When the loop finally closes, we can confidently prove that the biology the machine learning model predicted matches the physical reality we see in the lab. 🚀Follow along over the coming weeks as we continue this series, covering how Recursion maps biology, optimizes precision drug design, and closes the gap between patient and discovery. #TechBio #AIdrugdiscovery #AI #drugdiscovery

Recursion

1,014,580 次观看 • 1 个月前

We built a website to make it super easy to get started with protein design and participate in the Anthropic x Adaptyv protein design competition ▶️ This week, we're challenging people to design an Ozempic-like drug. You can get started in just a few clicks, using the newly released RFD4 model to design proteins targeting the GLP-1 receptor. We'll synthesize and test 1000+ of these designs in our automated biology lab, including whether they actually activate the cell receptor, and make all the results publicly available

We built a website to make it super easy to get started with protein design and participate in the Anthropic x Adaptyv protein design competition ▶️ This week, we're challenging people to design an Ozempic-like drug. You can get started in just a few clicks, using the newly released RFD4 model to design proteins targeting the GLP-1 receptor. We'll synthesize and test 1000+ of these designs in our automated biology lab, including whether they actually activate the cell receptor, and make all the results publicly available

Julian Englert

23,644 次观看 • 17 天前

Today marks a pivotal moment for Isomorphic Labs. We have secured $2.1 Billion in our second external funding round, led by Thrive Capital. They are joined at the table by Alphabet, GV and new investors MGX, Temasek, CapitalG and the UK Sovereign AI Fund. This milestone accelerates our ability to build the pioneering novel AI models that power our AI drug design engine (IsoDDE) and deploy them at scale: delivering scientific breakthroughs with a precision previously thought impossible, accelerating and expanding our pipeline of therapeutic programs toward the clinic. All with the ultimate goal of delivering life-changing new medicines to patients. Moving forward, we will scale our drug candidate pipelines across multiple therapeutic areas, expand our global footprint, and push the boundaries of frontier AI research to power our drug design engine. Deeply grateful to everyone sharing our vision to solve all disease with AI. Let’s build the future of medicine. Read the full announcement here:

Today marks a pivotal moment for Isomorphic Labs. We have secured $2.1 Billion in our second external funding round, led by Thrive Capital. They are joined at the table by Alphabet, GV and new investors MGX, Temasek, CapitalG and the UK Sovereign AI Fund. This milestone accelerates our ability to build the pioneering novel AI models that power our AI drug design engine (IsoDDE) and deploy them at scale: delivering scientific breakthroughs with a precision previously thought impossible, accelerating and expanding our pipeline of therapeutic programs toward the clinic. All with the ultimate goal of delivering life-changing new medicines to patients. Moving forward, we will scale our drug candidate pipelines across multiple therapeutic areas, expand our global footprint, and push the boundaries of frontier AI research to power our drug design engine. Deeply grateful to everyone sharing our vision to solve all disease with AI. Let’s build the future of medicine. Read the full announcement here:

Isomorphic Labs

310,089 次观看 • 6 个月前

We launched the first Engineering World Model, SKY-001, this week and it already outperforms all the frontier LLMs in CAD and BIM in speed, cost and precision. Here’s an example of how our model compares to OpenAI’s Astra, Anthropic’s Fable 5.1 and Google’s Gemini 4. These results did not happen by accident. Read our technical blog to find out how we built SKY-001 (🔗 below).

We launched the first Engineering World Model, SKY-001, this week and it already outperforms all the frontier LLMs in CAD and BIM in speed, cost and precision. Here’s an example of how our model compares to OpenAI’s Astra, Anthropic’s Fable 5.1 and Google’s Gemini 4. These results did not happen by accident. Read our technical blog to find out how we built SKY-001 (🔗 below).

Skyfall AI

64,413 次观看 • 19 天前

Check out 'Inside Iso', our new video series taking people inside Isomorphic Labs, and learn what keeps our 'Isos' intrigued as we work to change the way drug discovery is done. Join us! #AI #machinelearning #drugdiscovery

Check out 'Inside Iso', our new video series taking people inside Isomorphic Labs, and learn what keeps our 'Isos' intrigued as we work to change the way drug discovery is done. Join us! #AI #machinelearning #drugdiscovery

Isomorphic Labs

30,389 次观看 • 3 年前