04.10-11.24-HST-Jason-Yim

Conference Video|Duration: 25:51
April 10, 2024
View this past event
Please login to view this video.
  • Video details
    Deep learning has enabled considerable progress in protein structure prediction and design. In particular, the same technology underlying ChatGPT and DALL-E is rapidly being integrated into the workflows for de novo protein design. I will describe our work on this recent progress, RoseTTAFold Diffusion (RFdiffusion), that has enabled a generative AI framework for a wide range of protein design challenges. At the same time, generative AI is far surpassing the capabilities of traditional methods. These challenges span designing binder, enzymes, and peptides that have therapeutic applications. I will discuss the innovations, challenges, and future outlook of how generative AI will enable advancements in protein design.
Locked Interactive transcript
Please login to view this video.
  • Video details
    Deep learning has enabled considerable progress in protein structure prediction and design. In particular, the same technology underlying ChatGPT and DALL-E is rapidly being integrated into the workflows for de novo protein design. I will describe our work on this recent progress, RoseTTAFold Diffusion (RFdiffusion), that has enabled a generative AI framework for a wide range of protein design challenges. At the same time, generative AI is far surpassing the capabilities of traditional methods. These challenges span designing binder, enzymes, and peptides that have therapeutic applications. I will discuss the innovations, challenges, and future outlook of how generative AI will enable advancements in protein design.
Locked Interactive transcript

More Videos From This Event

See all videos