Cradle

I'll be heading to JPM next week and I'm excited to be meeting a lot of amazing folks. If you want to make your therapeutics go vroom and not pay millions in royalties hit me up!

Today, we're announcing Research Portfolios. The place to showcase your best lab work. And that other lab work too, that never got it's day in the sun. Your lab notebook is valuable. Contained within it, amongst the scribbles, the coffee stains, and the master mix calculations, are pages and pages of learning. Research Portfolios are your space to share those learnings, whether they made it into a paper or they didn’t. Don’t know what I’m on about? To explain, we’ll be showing you a few Briefly Bio portfolios from our users. You’ll get an insight into their science, and maybe some ideas about aspects of your own research you could share. Today, we’re kicking off with Cradle. Cradle are simplifying protein engineering. Their AI models suggest better sequences for functional objectives. Their in-house lab builds and tests new generations of protein variants, and the data feeds back and improves the next round of predictions. Cradle are doing things differently, in so many ways. Not only in their tech, but also in their values. That includes publicly sharing the lessons they’ve learned. With that in mind, Cradle pulled back the curtain on two of their core wet lab processes, so that others might use and learn from them, sharing their protocols on their Briefly Portfolio: https://lnkd.in/eVhEHsXm. This includes all the details of their high throughput protein expression platform –  how they use Enpresso media to produce consistent yields at low volumes, as well as their experience with a mutagenesis technique called Darwin Assembly from Chris Cozens and Vitor Pinheiro. Jinel Shah at Cradle reflected that, “Something as simple as providing this workflow makes it more accessible. We’re trying to change the mentality because a lot of companies think that getting fancy equipment will solve all of their problems,” Cost-effectiveness, practicality, and transparency. Celebrating these qualities in science will help us all move forward faster. Check out their portfolio: https://lnkd.in/eVhEHsXm. #openscience #research #proteinengineering #cradle

Last week, Cradle announced their $73M Series B to further expand their AI-enabled protein engineering platform. So far, they’ve partnered with 21 customers and achieved cost reductions of up to 90% on R&D projects. On this week’s Vital Signs, I sat down with Cradle’s Co-Founder & CEO, Stef van Grieken. We discuss how Cradle plans to use the new funding, what’s been happening in AI x bio, the technical underpinnings of Cradle’s platform, and more. Some highlights: 🧪 The current state & future of AI x bio Stef says that we’re still in the early stages of AI x bio. He highlights three areas where AI can play an important role: 1) hit identification for easy targets via novel binders 2) structural de novo models where researchers have some understanding of the target but want to generate some variance 3) multi-property optimization to learn from experimental results and reduce the number of experimental cycles needed. Stef advocates for more experimental context in models and better benchmarks that are relevant to bio. He also talks about how the valuable datasets in bio are kept private, meaning the public datasets are inherently less valuable, which might bias models towards irrelevant directions.  🧪 How Cradle designs their models Stef shares how their foundation model has two major components: 1) A predictor component which has some knowledge of properties (e.g., stability, expression), works decently well in zero-shot, and sees all of the assay data. 2) A generator to search the local search space and is conditioned to understand the relevant domain (e.g., providing evolutionary information, providing some labeled data without leaking too much). Stef mentions that it’s easy to go out of domain in biology given the sparsity of data, so Cradle has invested significantly in model confidence around generated sequences.  🧪 What’s next for Cradle Stef explains Cradle’s three goals: 1) Most protein models to date are assuming a fixed vocabulary with natural amino acids, but Cradle wants to also represent non-natural ones. 2) Models are currently effective at optimizing a protein construct once a user has properly formatted it. Stef hopes to have models reformat from large libraries – e.g., generating a bispecific antibody instead of configuring from the individual components. 3) Cradle plans to do more zero-shot learning on their panel of assays in areas like immunogenicity. An insightful discussion on all things AI x bio! Listen to the full episode below: Spotify: https://bit.ly/3D0Cy20 Apple: https://bit.ly/3ZiSyDZ