Skip navigation
Proteins can now talk.
0:00 / 0:25

Proteins can now talk.

Adib

@adibvafa3.6K followers

Visit on X
1,566 likesDownloadOriginal post

207,161 views

Proteins can now talk. Introducing BioReason-Pro, the first reasoning model for protein function. A thread🧵

71 Comments

Comments from the original post on X (Twitter)

  • Adib's profile picture

    @adibvafa

    BioReason-Pro is a multimodal LLM that brings protein foundation models and LLMs together for reasoning.

    48 likes

  • Adib's profile picture

    @adibvafa

    It was trained on 130K+ protein reasoning traces and then refined further with RL.

    27 likes

  • Adib's profile picture

    @adibvafa

    It outperforms all prior methods in both Gene Ontology and free text prediction.

    20 likes

  • Adib's profile picture

    @adibvafa

    Even human experts preferred it over UniProt ground truth in 79% of the cases

    15 likes

  • Adib's profile picture

    @adibvafa

    BioReason-Pro correctly predicted a functional protein partner that was validated in a cryo-EM study. It's attention was right at the contact residues.

    21 likes

  • Adib's profile picture

    @adibvafa

    It has learned structural reasoning purely from trainnig. In a shocking case, when predicting protein scaffold activity, it attended to exactly the 3 residues out of thousands that had been repurposed from catalytic to scaffolding function.

    16 likes

  • Adib's profile picture

    @adibvafa

    You can talk to it here! Paper: Code: Data: Weights: , Catalogue of 240,000+ predictions:

    77 likes

  • Adib's profile picture

    @adibvafa

    This has been an incredible work of a big and powerful team! Thank you to Arman Seyed-Ahmadi (@arman1sa), Parsa Idehpour (@Radii2323), Omar Ibrahim, Purav Gupta, Jack Naimer, Kevin Zhu, Arnav Shah, Shihao Ma, Abhinav Adduri, Talu Güloglu, Nuo Liu, Haotian Cui, Arihant Jain, Max de Castro, Amirfaham Fallahpour, Antonio Cembellin-Prieto, John S. Stiles, Filip Nemčko, Alexander A. Nevue, Hyungseok C. Moon, Lucas Sosnick, Olivia Markham, Haonan Duan, Michelle Y. Y. Lee, Andrea F. M. Salvador, Chris J. Maddison, Christoph A. Thaiss, Chiara Ricci-Tam, Brian S. Plosky, Dave P. Burke (@davey_burke), Patrick D. Hsu (@pdhsu), Hani Goodarzi (@genophoria), and Bo Wang (@BoWang87). across Arc Institute (@arcinstitute), University Health Network (@UHN), Vector Institute (@VectorInst), University of Toronto (@UofT), Stanford University (@Stanford), and more!

    23 likes

  • Andrew White 🐦‍⬛'s profile picture

    @andrewwhite01

    Great project! I've been looking for someone to try this on protein function and your team did a great job!

    9 likes

  • Adib's profile picture

    @adibvafa

    Thank you Andrew, appreciate it!

  • rajan agarwal's profile picture

    @_rajanagarwal

    the goat strikes again

    6 likes

  • Adib's profile picture

    @adibvafa

    🫡

  • Adib's profile picture

    @adibvafa

    summarize the paper

    5 likes

  • Adib's profile picture

    @adibvafa

    LMFAO

  • Talu Güloglu's profile picture

    @gueloglut

    @adibvafa probably one of the most fascinating young researchers in ai x bio

    4 likes

  • Adib's profile picture

    @adibvafa

    🫡

    1 like

  • faraz's profile picture

    @farazdotai

    Great work Adib, very interesting!

    3 likes

  • Adib's profile picture

    @adibvafa

    thank you Faraz!!

    1 like

  • OpenMed's profile picture

    @OpenMed_AI

    Thanks for dropping the code and the weights

    2 likes

  • Adib's profile picture

    @adibvafa

    of course!

    1 like