PulseAugur
EN
LIVE 23:10:46

AI models advance antibody design with new contact-reasoning architectures

By PulseAugur Editorial · [4 sources] ·

Three new research papers introduce advanced AI architectures for antibody complementarity-determining region (CDR) design. ConTact and AgForce tackle the challenge of explicitly reasoning about CDR-antigen contacts, improving structural quality and epitope awareness. EvoStruct bridges protein language models with equivariant graph neural networks, addressing vocabulary collapse and enhancing sequence recovery by leveraging evolutionary data. AI

IMPACT These new models offer improved accuracy and diversity in antibody design, potentially accelerating drug discovery and therapeutic development.

RANK_REASON Multiple research papers published on arXiv detailing new AI models for antibody design.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 4 sources. How we write summaries →

COVERAGE [4]

  1. arXiv cs.LG TIER_1 English(EN) · Mansoor Ahmed, Spencer VonBank, Nadeem Taj, Sujin Lee, Naila Jan, Murray Patterson ·

    ConTact: Contact-First Antibody CDR Design via Explicit Interface Reasoning

    arXiv:2605.21600v1 Announce Type: new Abstract: Computational antibody CDR design methods condition on antigen structure to generate binding loops, yet existing architectures conflate two fundamentally distinct sub-problems: identifying which CDR positions will contact the antige…

  2. arXiv cs.LG TIER_1 English(EN) · Mansoor Ahmed, Murray Patterson ·

    AgForce Enables Antigen-conditioned Generative Antibody Design

    arXiv:2605.21610v1 Announce Type: new Abstract: Antibody design methods condition on antigen structure to generate complementarity-determining regions (CDR), yet a systematic evaluation of baseline methods reveals that they largely ignore the antigen input. We identify three fail…

  3. arXiv cs.LG TIER_1 English(EN) · Murray Patterson ·

    EvoStruct: Bridging Evolutionary and Structural Priors for Antibody CDR Design via Protein Language Model Adaptation

    Equivariant graph neural network (GNN) methods for antibody complementarity-determining region (CDR) design achieve the highest sequence recovery but suffer from severe vocabulary collapse. The current best GNN methods over-predict very few amino acids, such as tyrosine and glyci…

  4. Hugging Face Daily Papers TIER_1 English(EN) ·

    EvoStruct: Bridging Evolutionary and Structural Priors for Antibody CDR Design via Protein Language Model Adaptation

    Equivariant graph neural network (GNN) methods for antibody complementarity-determining region (CDR) design achieve the highest sequence recovery but suffer from severe vocabulary collapse. The current best GNN methods over-predict very few amino acids, such as tyrosine and glyci…

RELATED ENTITIES

RELATED TOPICS