Peptide sequence tag identification method based on graph annotation mechanism
CN116825198B · Hunan University of Technology and Business · priority 2023
| Patent | CN116825198B ↗ |
| Status | Active |
| Priority date | 2023-07-14 |
| Filed | 2023-07-14 |
| Published (granted) | 2024-05-10 |
| Assignee | Hunan University of Technology and Business (printed Hunan University of Technology) |
| Also published as | CN116825198A |
Abstract
Translated from Chinese 本公开实施例中提供了一种基于图注意机制的肽序列标签鉴定方法，属于生物信息学技术领域，具体包括：对二级质谱图进行预处理，并基于图论构建谱峰连接图生成候选肽序列标签集；利用标签注意力模块构建标签嵌入矩阵，从训练数据中提取候选序列标签集中标签之间的顺序相关性；通过卷积神经网络模块进行特征提取，并将特征作为通道注意力模块的输入提取蛋白质序列的组成相关性；将标签注意力模块和通道注意力模块融合，构建完整的标签分类模型，从而生成肽序列标签；将肽序列标签在蛋白质数据库中进行搜索得到理论质谱图，并其与实验质谱图进行匹配打分，得到二级质谱图对应的蛋白质序列。通过本公开的方案，提高了计算效率、精准度和适应性。 The disclosed embodiment provides a peptide sequence tag identification method based on a graph attention mechanism, which belongs to the field of bioinformatics technology, and specifically includes: preprocessing the secondary mass spectrometer, and constructing a peak connection graph based on graph theory to generate a candidate peptide sequence tag set; using a tag attention module to construct a tag embedding matrix, and extracting the order correlation between tags in the candidate sequence tag set from the training data; extracting features through a convolutional neural network module, and using the features as the input of a channel attention module to extract the composition correlation of the protein sequence; fusing the tag attention module and the channel attention module to construct a complete tag classification model, thereby generating peptide sequence tags; searching the peptide sequence tags in a protein database to obtain a theoretical mass spectrum, and matching and scoring it with the experimental mass spectrum to obtain the protein sequence corresponding to the secondary mass spectrum. Through the scheme disclosed in the present invention, the computational efficiency, accuracy and adaptability are improved.
Why it is here
Hunan University of Technology and Business: peptide sequence tags from a spectrum peak graph, ranked with an attention network.
Inventors (6)
李闯, 刘纯洋, 李小龙, 何典, 黄华, 王和诗
An inventor links to an author page only where the same person is evident, never on a name alone.
Catalogued papers it cites (1)
- SWPepNovo: An Efficient De Novo Peptide Sequencing Tool for Large-scale MS/MS Spectra Analysis (2019) added by the examiner
Matched from the patent’s 5 literature citations by the rules the paper citation graph uses.
Catalogued patents it cites (4)
- US11573239B2 Methods and systems for de novo peptide sequencing using deep learning
- US11644470B2 Systems and methods for de novo peptide sequencing using deep learning and spectrum pairs
- CN113921086A Protein de novo peptide sequencing method and system based on mass spectrometry
- CN103852513B A kind of peptide section de novo sequencing method and system based on HCD and ETD mass spectrogram
Cited by (1 patent families, 1 organisations)
- 山东鲁润阿胶药业有限公司 company: CN117095743B
Every family, of any applicant, that cites this one, as Google Patents lists them; an assignee is given its English name.