相关实验视频

Updated: Sep 15, 2025

Determining the Likelihood of Variant Pathogenicity Using Amino Acid-level Signal-to-Noise Analysis of Genetic Variation

Determining the Likelihood of Variant Pathogenicity Using Amino Acid-level Signal-to-Noise Analysis of Genetic Variation

Published on: January 16, 2019

通过使用预先训练的生物LLM来预测人类跨膜蛋白和相关疾病中的病原性突变

Lexin Cao¹, Lijun Quan^2,3,4, Qiufeng Chen¹

¹School of Computer Science and Technology, Soochow University, Suzhou, China.

Communications biology

|July 15, 2025

概括

这项研究介绍了MutDPAL,这是一种用于识别膜蛋白中的病原性突变的新型深度学习工具. MutDPAL准确地预测了突变效应,并将它们分为15种疾病类别,优于现有的方法.

科学领域:

基因组学就是基因组学.
计算生物学计算生物学
生物化学生物化学

背景情况:

膜蛋白中的错误突变可以通过改变蛋白质结构和功能来引起疾病.
目前的计算工具缺乏对膜蛋白和细粒度疾病分类的特异性.

研究的目的:

开发 MutDPAL,这是一种用于准确预测膜蛋白突变病原性的深度学习方法.
为了使细粒度,多标签的突变分类到特定的疾病类别.

主要方法:

用了两个预训练的生物大语言模型 (Bio-LLMs) 来测序和跨膜环境特征.
在膜蛋白中使用交叉注意力机制进行疾病蛋白关联学习.

主要成果:

与现有方法相比,MutDPAL在预测膜蛋白突变致病性方面取得了更好的表现.
在15个不同的疾病类别的多标签分类中表现出高准确性.

结论:

MutDPAL是第一个整合跨膜环境和疾病编码的方法,用于精确的突变分类.
提供了关于膜蛋白和相关疾病的误解突变致病性和相关疾病的宝贵见解.

更多相关视频

In Vivo Functional Study of Disease-associated Rare Human Variants Using Drosophila

In Vivo Functional Study of Disease-associated Rare Human Variants Using Drosophila

Published on: August 20, 2019

In Vivo Modeling of the Morbid Human Genome using Danio rerio

In Vivo Modeling of the Morbid Human Genome using Danio rerio

Published on: August 24, 2013

相关概念视频

Single-pass Transmembrane Proteins

Single-pass Transmembrane Proteins

Integral membrane proteins are tightly associated with the cell membrane and play a crucial role in cell communication, signaling, adhesion, and transport of the molecules. Some integral membrane proteins are present only in the membrane monolayer. For example, the enzyme fatty acid amide hydrolase is present in the cytoplasmic side of the membrane monolayer. In contrast, another type of integral membrane protein, also known as a transmembrane protein, spans across the membrane. Transmembrane...