Miles Gander

@milesgander.bsky.social

Head of Large Libraries @ http://Cradle.Bio - Engineering better proteins, faster. SynBio+AI/ML - Synthetic biologist and Seattle sports fan. https://www.linkedin.com/in/mileswgander

Excited to announce I've joined www.cradle.bio as Head of Large Libraries. The bottleneck in ML for protein engineering is data. Other domains train on massive existing datasets; in biology, every data point is generated experimentally and doing that well at scale is challenging.

AI is revolutionizing an incredibly broad range of sectors. Enjoyed the chance to share how Birch Biosciences is using AI to power our enzymatic recycling technology. Check out more details below!

Bild

Excited to share Birch Biosciences' progress towards a circular plastics economy at this year's SIMB conference! If anyone in my network is heading to the 2025 Society for Industrial Microbiology and Biotechnology meeting in SF next week, let's connect!

Bild

Big news! Birch has entered into a global patent license agreement with NREL! This agreement will drive commercialization of our plastic recycling tech by combining NREL innovation and our AI-powered enzymatic platform. A big step toward circularizing plastics! www.businesswire.com/news/home/20...

Birch Biosciences Enters into Global Patent License Agreement with NREL to Advance Enzymatic Plastic Recycling Process

Birch Biosciences, a startup focused on AI-powered enzymatic plastic recycling, today announced a global patent license agreement with the U.S. Department of...

businesswire.com

We compared the calibration of various machine learning uncertainty estimation methods for protein engineering. No method excels across all scenarios, and uncertainty-based strategies for optimization often did not outperform methods without uncertainty.

Approach, datasets, and tasksMiscalibration area vs. root mean square error (RMSE) Active learningBayesian optimization