Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2024 May 20;25(3):bbae247. doi: 10.1093/bib/bbae247

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2024. Published by Oxford University Press.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

PMC Copyright notice

Overview of UltraMutate’s architecture and workflow. (A) For a given pHLA pair, the SL policy network takes embedded sequences and flattened vectors from three distinct attention matrices as input. It uses a linear layer to combine features from four input heads to generate a conditional probability distribution for all possible mutation actions. (B) The interactive environment for the agent, where rewards are given based on the binding probability of the mutated peptide with the HLA molecule and the homology between the mutated and original peptides. (C) UltraMutate’s overall workflow includes: (1) supervised learning with the AOMP dataset; (2) fine-tuning the policy network using RL with PPO; and (3) integrating with an MCTS algorithm to determine the final mutation action.

Inline graphic — Overview of UltraMutate’s architecture and workflow. (A) For a given pHLA pair, the SL policy network takes embedded sequences and flattened vectors from three distinct attention matrices as input. It uses a linear layer to combine features from four input heads to generate a conditional probability distribution for all possible mutation actions. (B) The interactive environment for the agent, where rewards are given based on the binding probability of the mutated peptide with the HLA molecule and the homology between the mutated and original peptides. (C) UltraMutate’s overall workflow includes: (1) supervised learning with the AOMP dataset; (2) fine-tuning the policy network using RL with PPO; and (3) integrating with an MCTS algorithm to determine the final mutation action.