Protein Structure Prediction in the 3D HP Model Using Deep Reinforcement Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Espitia, Giovanny, Pang, Yui Tik, Gumbart, James C.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915085468827648
author Espitia, Giovanny
Pang, Yui Tik
Gumbart, James C.
author_facet Espitia, Giovanny
Pang, Yui Tik
Gumbart, James C.
contents We address protein structure prediction in the 3D Hydrophobic-Polar lattice model through two novel deep learning architectures. For proteins under 36 residues, our hybrid reservoir-based model combines fixed random projections with trainable deep layers, achieving optimal conformations with 25% fewer training episodes. For longer sequences, we employ a long short-term memory network with multi-headed attention, matching best-known energy values. Both architectures leverage a stabilized Deep Q-Learning framework with experience replay and target networks, demonstrating consistent achievement of optimal conformations while significantly improving training efficiency compared to existing methods.
format Preprint
id arxiv_https___arxiv_org_abs_2412_20329
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Protein Structure Prediction in the 3D HP Model Using Deep Reinforcement Learning
Espitia, Giovanny
Pang, Yui Tik
Gumbart, James C.
Machine Learning
Artificial Intelligence
Biomolecules
We address protein structure prediction in the 3D Hydrophobic-Polar lattice model through two novel deep learning architectures. For proteins under 36 residues, our hybrid reservoir-based model combines fixed random projections with trainable deep layers, achieving optimal conformations with 25% fewer training episodes. For longer sequences, we employ a long short-term memory network with multi-headed attention, matching best-known energy values. Both architectures leverage a stabilized Deep Q-Learning framework with experience replay and target networks, demonstrating consistent achievement of optimal conformations while significantly improving training efficiency compared to existing methods.
title Protein Structure Prediction in the 3D HP Model Using Deep Reinforcement Learning
topic Machine Learning
Artificial Intelligence
Biomolecules
url https://arxiv.org/abs/2412.20329