CFP-Gen: Combinatorial Functional Protein Generation via Diffusion Language Models

TL;DR

CFP-Gen is a diffusion language model that enables the design of novel, multifunctional proteins by integrating multiple constraints across functional, sequence, and structural modalities, improving control and diversity in protein generation.

Contribution

The paper introduces CFP-Gen, a novel multimodal diffusion model with modules for functional annotation guidance and residue-level control, advancing de novo protein design capabilities.

Findings

High-throughput generation of functional proteins

Achieves high success rate in designing multifunctional proteins

Generates proteins with functionality comparable to natural ones

Abstract

Existing PLMs generate protein sequences based on a single-condition constraint from a specific modality, struggling to simultaneously satisfy multiple constraints across different modalities. In this work, we introduce CFP-Gen, a novel diffusion language model for Combinatorial Functional Protein GENeration. CFP-Gen facilitates the de novo protein design by integrating multimodal conditions with functional, sequence, and structural constraints. Specifically, an Annotation-Guided Feature Modulation (AGFM) module is introduced to dynamically adjust the protein feature distribution based on composable functional annotations, e.g., GO terms, IPR domains and EC numbers. Meanwhile, the Residue-Controlled Functional Encoding (RCFE) module captures residue-wise interaction to ensure more precise control. Additionally, off-the-shelf 3D structure encoders can be seamlessly integrated to impose geometric constraints. We demonstrate that CFP-Gen enables high-throughput generation of novel proteins with functionality comparable to natural proteins, while achieving a high success rate in designing multifunctional proteins. Code and data available at https://github.com/yinjunbo/cfpgen.