[Scale ML] Jeremy Wohlwend: Biomolecular Modeling with Boltz-1

Time: 4 pm Wednesday (April 9th)

Location: 45-792 (the big glass room in the middle on the 7th floor)

Zoom: https://mit.zoom.us/j/91697262920 (password: mitmlscale)

Speaker: Jeremy Wohlwend

Title: Biomolecular Modeling with Boltz-1

Asbtract: Understanding biomolecular interactions is fundamental to advancing fields like drug discovery and protein design. In this talk, we introduce Boltz-1, an open-source deep learning model incorporating innovations in model architecture, speed optimization, and data processing achieving AlphaFold3-level accuracy in predicting the 3D structures of biomolecular complexes. Boltz-1 demonstrates a performance on-par with state-of-the-art commercial models on a range of diverse benchmarks, setting a new benchmark for commercially accessible tools in structural biology. By releasing the training and inference code, model weights, datasets, and benchmarks under the MIT open license, we aim to foster global collaboration, accelerate discoveries, and provide a robust platform for advancing biomolecular modeling.

Bio: I’m a PhD student at MIT in Regina Barzilay’s group, where I focus on biomolecular modeling. Previously, I worked as a staff research engineer at ASAPP Inc., leading a small team dedicated to enhancing the efficiency and scalability of language models. Today, my research spans foundation models for biology and applied work in immunology.

Scale ML: We are a cross-lab graduate student collective focusing on Algorithms That Learn and Scale. We currently host bi-weekly seminars and will have hands on sessions and research socials in the future. We are funded by generous donations from Pulkit Agrawal, Yoon Kim and BVP.