Abstract:Designing extit{de novo} proteins beyond those found in nature holds significant promise for advancements in both scientific and engineering applications. Current methodologies for protein design often rely on AI-based models, such as surrogate models that address end-to-end problems by linking protein structure to material properties or vice versa. However, these models frequently focus on specific material objectives or structural properties, limiting their flexibility when incorporating out-of-domain knowledge into the design process or comprehensive data analysis is required. In this study, we introduce ProtAgents, a platform for extit{de novo} protein design based on Large Language Models (LLMs), where multiple AI agents with distinct capabilities collaboratively address complex tasks within a dynamic environment. The versatility in agent development allows for expertise in diverse domains, including knowledge retrieval, protein structure analysis, physics-based simulations, and results analysis. The dynamic collaboration between agents, empowered by LLMs, provides a versatile approach to tackling protein design and analysis problems, as demonstrated through diverse examples in this study. The problems of interest encompass designing new proteins, analyzing protein structures and obtaining new first-principles data -- natural vibrational frequencies -- via physics simulations. The concerted effort of the system allows for powerful automated and synergistic design of extit{de novo} proteins with targeted mechanical properties. The flexibility in designing the agents, on one hand, and their capacity in autonomous collaboration through the dynamic LLM-based multi-agent environment on the other hand, unleashes great potentials of LLMs in addressing multi-objective materials problems and opens up new avenues for autonomous materials discovery and design.

Automating Exploratory Proteomics Research via Language Models

Automation to Enable High-Throughput Chemical Proteomics

A fully automated system with online sample loading, isotope dimethyl labeling and multidimensional separation for high-throughput quantitative proteome analysis.

AutoProteinEngine: A Large Language Model Driven Agent Framework for Multimodal AutoML in Protein Engineering

ProtTrans: Toward Understanding the Language of Life Through Self-Supervised Learning

ProteusAI: An Open-Source and User-Friendly Platform for Machine Learning-Guided Protein Design and Engineering

Comparative Performance Evaluation of Large Language Models for Extracting Molecular Interactions and Pathway Knowledge

Decoding proteome functional information in model organisms using protein language models.

Proteomics for validation of automated gene model predictions.

From Intention To Implementation: Automating Biomedical Research via LLMs

ProtTrans: Towards Cracking the Language of Life's Code Through Self-Supervised Deep Learning and High Performance Computing

Unifying Sequences, Structures, and Descriptions for Any-to-Any Protein Generation with the Large Multimodal Model HelixProtX

ProtAgents: Protein discovery via large language model multi-agent collaborations combining physics and machine learning

Synergizing Human Expertise and AI Efficiency with Language Model for Microscopy Operation and Automated Experiment Design

Improving the success rate of proteome analysis by modeling protein-abundance distributions and experimental designs

Automating literature screening and curation with applications to computational neuroscience

Discovering Patterns to Extract Protein-Protein Interactions from the Literature: Part Ii

AnnoPRO: an Innovative Strategy for Protein Function Annotation Based on Image-like Protein Representation and Multimodal Deep Learning

Democratizing protein language models with parameter-efficient fine-tuning

The Future of Large-Scale Collaborative Proteomics

A rapid methods development workflow for high-throughput quantitative proteomic applications