PatchProt:使用蛋白质基础模型进行疏水性补丁预测
Dea Gogishvili1,2, Emmanuel Minois-Genin1, Jan van Eck2
1Bioinformatics, Computer Science Department, Vrije Universiteit Amsterdam, Amsterdam, 1081 HV, The Netherlands.
Bioinformatics advances
|November 1, 2024
概括
预测蛋白质疏水性补丁是具有挑战性的. 用多任务学习微调大型语言模型可以提高蛋白质表面可访问性和二次结构预测的准确性.
科学领域:
- 计算生物学是一种计算生物学.
- 蛋白质结构预测 蛋白质结构预测
- 在生物信息学中的机器学习.
背景情况:
- 蛋白质表面的疏水斑对于相互作用至关重要,并与疾病有关.
- 从蛋白质序列中预测这些补丁是一个重要的计算挑战.
- 基础模型和多任务深度学习为数据缺口和改进的预测提供了潜在的解决方案.
研究的目的:
- 开发一种新的方法来预测蛋白质表面暴露的疏水性斑块.
- 利用领先的大型语言模型 (进化规模模型 - ESM-2) 和参数高效微调.
- 通过结合相关的本地和全球预测任务来增强模型表示.
主要方法:
- 使用了进化规模模型 (ESM-2) 基础模型.
- 采用一个参数有效的微调方法,以实现高效的模型训练.
- 集成的多任务深度学习,对本地 (残留) 和全球 (蛋白质) 任务的培训.
主要成果:
- 开发了PatchProt,这是一个准确预测疏水性补丁面积的模型.
- 在预测二级结构和表面可访问性等主要任务方面,PatchProt的性能优于现有的方法.
- 对相关的本地任务的培训可以明显改善对更复杂的全球任务的预测.
结论:
- 微调基础模型与多任务学习对于蛋白质性质预测非常有效.
- PatchProt为基于序列的蛋白质性质预测设定了一个新的基准.
- 这种方法突出了通过相关任务培训来丰富模型表示的潜力.
相关概念视频
01:19Fluid Mosaic Model
11.5K
Scientists identified the plasma membrane in the 1890s and its principal chemical components (lipids and proteins) by 1915. The model for plasma membrane structure, proposed in 1935 by Hugh Davson and James Danielli, was the first model to be widely accepted in the scientific community. The model was based on the plasma membrane's "railroad track" appearance in early electron micrographs. Davson and Danielli theorized that the plasma membrane's structure resembled a sandwich...
11.5K
01:22Protein Folding
117.4K
Overview
117.4K
01:49Conserved Binding Sites
4.2K
Many proteins’ biological role depends on their interactions with their ligands, small molecules that bind to specific locations on the protein known as ligand-binding sites. Ligand-binding sites are often conserved among homologous proteins as these sites are critical for protein function.
Binding sites are often located in large pockets, and if their location on a protein’s surface is unknown, it can be predicted using various approaches. The energetic method computationally...
Binding sites are often located in large pockets, and if their location on a protein’s surface is unknown, it can be predicted using various approaches. The energetic method computationally...
4.2K
02:04Protein-protein Interfaces
12.5K
Many proteins form complexes to carry out their functions, making protein-protein interactions (PPIs) essential for an organism's survival. Most PPIs are stabilized by numerous weak noncovalent chemical forces. The physical shape of the interfaces determines the way two proteins interact. Many globular proteins have closely-matching shapes on their surfaces, which form a large number of weak bonds. Additionally, many PPIs occur between two helices or between a surface cleft and a...
12.5K
01:26Membrane Fluidity
11.0K
Membrane fluidity is explained by the fluid mosaic model of the cell membrane, which describes the plasma membrane structure as a mosaic of components—including phospholipids, cholesterol, proteins, and carbohydrates—that gives the membrane a fluid character.
Mosaic nature of the membrane
The mosaic characteristic of the membrane helps the plasma membrane remain fluid. The integral proteins and lipids exist as separate but loosely-attached molecules in the membrane. The membrane is...
Mosaic nature of the membrane
The mosaic characteristic of the membrane helps the plasma membrane remain fluid. The integral proteins and lipids exist as separate but loosely-attached molecules in the membrane. The membrane is...
11.0K
01:25Aquaporins
4.8K
Aquaporins or AQPs are a family of integral membrane proteins whose primary function is to transport water, while some called aquaglyceroporins also transport glycerol. In addition, aquaporins have also been suspected to be involved in transporting volatile substances, such as carbon dioxide and ammonia, across membranes. Such AQPs that act as gas channels are often highly expressed in cells involved in the gaseous exchange, such as red blood cells, epithelial cells, and pulmonary capillaries.
4.8K