Shane Lofgren retweeted
Really great collaboration between @ManifoldBio and @nvidia to test 1M de novo designed binders against 127 targets, measuring over 100 million potential protein-protein interactions! This was a great collaboration with some very exciting results. NVIDIA's new Proteina-Complexa method is SOTA for de novo minibinder design. If you're interested in designing minibinders to targets you couldn't hit with other methods, try it out!
I'm particularly excited about what this large scale data enables. As we generate 1000s of experimentally validated structures, this data becomes the input to training new protein design models. At Manifold, we are generating datasets of this size continuously, and have experimentally validated thousands of de novo designed binders across many formats (VHH, minibinders, peptides, etc). New models will open up new hard to hit targets, paired with our large scale in vivo measurement, will enable us to create previously impossible therapeutics.
Up next is training new models on this and other data we have generated, stay tuned! And thanks to NVIDIA for setting up such a great collaboration, its been fun and fruitful!
🧵 We ran the largest head-to-head benchmark of protein binder design methods in the wet lab.
Project page: research.nvidia.com/labs/gen…
1 million designs. 127 targets. RFdiffusion, BindCraft, BoltzGen, and Proteina-Complexa — all tested side by side.👇
Shane Lofgren retweeted
What you’re seeing at @ManifoldBio is what happens when two exponentials begin to resonate–the evolution of AI and the evolution of massively parallel molecular measurement and design.
When we started Manifold Bio with @PierceOgdenJ , @geochurch , and @slofgren, we set out to do something bold: to build an AI-first biopharma company that could map the astronomic space of molecular design against the complexity of human biology.
We had already become world leaders in the secret art of combining Machine Learning and Multiplexed Libraries, or ML². We believed this approach could be transformative to drug discovery.
AI in drug discovery has always faced one fundamental barrier: the translational gap between what works in a petri dish and what will work in a patient. We believed that closing that gap required a way to power AI models with massive-scale experimental feedback directly from living organisms.
Such a platform would be positioned to solve grand challenges in human health, including creating medicines that could cross the blood-brain barrier.
To succeed in such a challenge, it not only takes a culture of scientific boldness, but also bold partners. The boldness of the Roche team was clear from the moment we started mapping out this collaboration, both in their rigor in evaluation of the Manifold platform and the ambition we both shared in how to best leverage our approach. This is the company that has literally integrated the Genentech DNA, and charted many firsts along the way, including some of the most compelling brain-shuttling clinical data to date. It’s a privilege to combine forces to advance the frontier of brain delivery together.
This deal is fantastic validation of the power of our approach. And we’re just getting started.
The hardest problems in medicine will not be solved by incremental innovation.
They will require a fundamentally new approach that grounds AI-driven design in large-scale data from living systems.
Excited to announce our partnership with @Roche
manifold.bio/news/manifold-b…
Shane Lofgren retweeted
We're stil writing the @ManifoldBio story, but @ElliotHershberg just dropped a fantastic piece capturing the journey so far on his blog Century of Biology.
Shane Lofgren retweeted
.@manifoldbio won the grand prize in this year's @Harvard President's Innovation Challenge. The transformative company builds technology that exponentially increases measurement throughput at all stages of drug development. https://nitter.cf/t.co/R0dgrL3oIh
Shane Lofgren retweeted
Excited to share our seed raise @ManifoldBio to build multiplexed tech for protein therapeutics dev with @PierceOgdenJ, @slofgren, @geochurch and a talented team! Huge support from @Playground_VC and @fiftyyears. Thanks @FierceBiotech for the nice piece! fiercebiotech.com/biotech/co…
Shane Lofgren retweeted
Replying to @slofgren
Multiple, but basically slows/prevents recruitment of bone marrow cells that establish tumor-friendly microenvironment, e.g., MDSC
Shane Lofgren retweeted
Replying to @slofgren
10-year data looks similar - with BioPharms topping the dispersion chart.










