Foundation Models for AI-Enabled Biological Design

TL;DR

This survey reviews recent advances in foundation models for AI-driven biological design, highlighting architectures, challenges, and future directions in applying large-scale models to protein, molecule, and genomic tasks.

Contribution

It provides a comprehensive taxonomy of current models and discusses key challenges and solutions for adapting foundation models to biological applications.

Findings

Identifies key architectures for biological sequence modeling

Discusses challenges in controllability and multi-modal integration

Outlines future research directions in biological foundation models

Abstract

This paper surveys foundation models for AI-enabled biological design, focusing on recent developments in applying large-scale, self-supervised models to tasks such as protein engineering, small molecule design, and genomic sequence design. Though this domain is evolving rapidly, this survey presents and discusses a taxonomy of current models and methods. The focus is on challenges and solutions in adapting these models for biological applications, including biological sequence modeling architectures, controllability in generation, and multi-modal integration. The survey concludes with a discussion of open problems and future directions, offering concrete next-steps to improve the quality of biological sequence generation.