蛋白质需要额外的关注:提高蛋白质语言模型在突变数据集上的预测能力,使用提示令牌
Xinning Li1, Ryann M Perez1, Sam Giannakoulias2
1Department of Chemistry, University of Pennsylvania, Philadelphia, PA 19104, United States.
NAR genomics and bioinformatics
|September 29, 2025
概括
暗示令牌学习 (HTL) 改善了蛋白质语言模型 (PLM) 在突变后对蛋白质功能的预测. 这种新的战略增强了PLM.
科学领域:
- 计算生物学是一种计算生物学.
- 生物信息学是一种生物信息学.
- 蛋白质工程是一种蛋白质工程.
背景情况:
- 预测突变后的蛋白质功能对于理解疾病和设计新型蛋白质至关重要.
- 蛋白质语言模型 (PLM) 是有前途的,但需要有效的微调策略来预测突变影响.
- 目前的方法可能无法完全捕捉突变对蛋白质功能的细微影响.
研究的目的:
- 开发和评估一种新的代币化策略,暗示代币学习 (HTL),用于微调PLM来预测突变后的蛋白质功能.
- 通过使用各种突变数据集,在不同的PLM架构和大小中评估HTL的性能.
- 研究HTL改善突变影响预测的机制.
主要方法:
- 微调四个预训练的蛋白质语言模型 (PLM),具有不同的架构和大小,使用新的提示令牌学习 (HTL) 策略.
- 在四个不同的蛋白质突变数据集上对HTL性能进行基准测试,评估加权F1得分的改善.
- 从精心调整的PLM中对嵌入的稀疏自动编码器进行训练,以分析隐藏空间表示.
主要成果:
- 在大多数情况下,当HTL应用于PLM时,观察到加权F1得分的显著改善.
- 潜伏空间的分析显示,在PLM训练后,功能蛋白域内的活性残留物数量增加了.
- 这些发现表明,HTL使PLM能够捕获更多生物相关的蛋白质表示.
结论:
- 提示令牌学习 (HTL) 是提高蛋白质语言模型 (PLM) 在突变后预测蛋白质功能的有效策略.
- 与HTL微调的PLM显示出更好的捕获与突变影响相关的生物相关信息的能力.
- 这项研究为PLM的解释性提供了宝贵的见解,并为推进蛋白质功能预测提供了有前途的方法.
更多相关视频
08:04Identification and Classification of Position-specific GABAA Receptor Subunit Missense Variants for Their Role In Hippocampal Pyramidal Neurons
Published on: June 6, 2025
1.4K
07:08Optimization of Synthetic Proteins: Identification of Interpositional Dependencies Indicating Structurally and/or Functionally Linked Residues
Published on: July 14, 2015
7.7K
相关概念视频
02:07Improving Translational Accuracy
14.1K
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
14.1K
02:07Improving Translational Accuracy
3.5K
3.5K
01:30Point and Frameshift Mutations
857
Point mutations are genetic alterations involving the change of a single nucleotide base pair in DNA. Depending on how the alteration affects protein synthesis, they can lead to various consequences.Point mutations fall into the following types:Silent mutations occur when a nucleotide change does not alter the amino acid sequence due to the redundancy of the genetic code. For instance, changing ACC to ACA still encodes threonine, leaving the protein function unaffected. This occurs because...
857
01:39Mutations
94.3K
Overview
94.3K
01:35Mutations
42.7K
Mutations are changes in the sequence of DNA. These changes can occur spontaneously or they can be induced by exposure to environmental factors. Mutations can be characterized in a number of different ways: whether and how they alter the amino acid sequence of the protein, whether they occur over a small or large area of DNA, and whether they occur in somatic cells or germline cells.
Chromosomal Alterations Are Large-Scale Mutations
While point mutations are changes in a single nucleotide in...
Chromosomal Alterations Are Large-Scale Mutations
While point mutations are changes in a single nucleotide in...
42.7K
02:28Leaky Scanning
5.6K
During most eukaryotic translation processes, the small 40S ribosome subunit scans an mRNA from its 5' end until it encounters the first start AUG codon. The large 60S ribosomal subunit then joins the smaller one to initiate protein synthesis. The location of the translation initiation is largely determined by the nucleotides near the start codon as there may be multiple translation initiation sites present on the mRNA. Marilyn Kozak discovered that the sequence RCCAUGG (where R...
5.6K