CoSpred: Machine Learning Workflow to Predict Tandem Mass Spectrum in Proteomics.

IF 3.9 4区 生物学 Q2 BIOCHEMICAL RESEARCH METHODS
Proteomics Pub Date : 2025-06-30 DOI:10.1002/pmic.70004
Liang Xue, Shivani Tiwary, Mykola Bordyuh, Robert Stanton
{"title":"CoSpred: Machine Learning Workflow to Predict Tandem Mass Spectrum in Proteomics.","authors":"Liang Xue, Shivani Tiwary, Mykola Bordyuh, Robert Stanton","doi":"10.1002/pmic.70004","DOIUrl":null,"url":null,"abstract":"<p><p>In mass spectrometry-based proteomics, the use of deep learning algorithms can help improve the identification rates of peptides and proteins through the generation of high-fidelity theoretical spectrum which can be used as the basis of a more complete spectral library than those presently available, especially for unobserved protein/genetic variants. Here we focus on providing an end-to-end user-friendly machine learning workflow, which we call Complete Spectrum Predictor (CoSpred). Using CoSpred users can create their own machine learning compatible training dataset and then train a machine learning model to predict both backbone and non-backbone ions. For the model a transformer encoder architecture is used to predict the complete MS/MS spectrum from a given peptide sequence. In addition to the transformer model provided in the package, the code is built modularly to allow for alternate ML models to be easily \"plugged in,\" allowing for spectrum prediction optimization given different experimental conditions. The CoSpred workflow (preprocessing→training→inference) provides a path for state-of-art ML capabilities to be more accessible to proteomics scientists.</p>","PeriodicalId":224,"journal":{"name":"Proteomics","volume":" ","pages":"e70004"},"PeriodicalIF":3.9000,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proteomics","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1002/pmic.70004","RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"BIOCHEMICAL RESEARCH METHODS","Score":null,"Total":0}
引用次数: 0

Abstract

In mass spectrometry-based proteomics, the use of deep learning algorithms can help improve the identification rates of peptides and proteins through the generation of high-fidelity theoretical spectrum which can be used as the basis of a more complete spectral library than those presently available, especially for unobserved protein/genetic variants. Here we focus on providing an end-to-end user-friendly machine learning workflow, which we call Complete Spectrum Predictor (CoSpred). Using CoSpred users can create their own machine learning compatible training dataset and then train a machine learning model to predict both backbone and non-backbone ions. For the model a transformer encoder architecture is used to predict the complete MS/MS spectrum from a given peptide sequence. In addition to the transformer model provided in the package, the code is built modularly to allow for alternate ML models to be easily "plugged in," allowing for spectrum prediction optimization given different experimental conditions. The CoSpred workflow (preprocessing→training→inference) provides a path for state-of-art ML capabilities to be more accessible to proteomics scientists.

CoSpred:预测蛋白质组学串联质谱的机器学习工作流程。
在基于质谱的蛋白质组学中,使用深度学习算法可以通过生成高保真理论谱来帮助提高肽和蛋白质的识别率,该理论谱可以作为比现有谱库更完整的谱库的基础,特别是对于未观察到的蛋白质/遗传变异。在这里,我们专注于提供端到端用户友好的机器学习工作流,我们称之为全谱预测器(CoSpred)。使用CoSpred,用户可以创建自己的机器学习兼容训练数据集,然后训练机器学习模型来预测主干和非主干离子。该模型采用了一种变压器编码器结构,从给定的肽序列中预测完整的质谱。除了包中提供的变压器模型外,代码是模块化构建的,允许轻松“插入”备用ML模型,从而允许在不同实验条件下进行频谱预测优化。CoSpred工作流程(预处理→训练→推理)为蛋白质组学科学家提供了一条更容易获得最先进的机器学习功能的途径。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Proteomics
Proteomics 生物-生化研究方法
CiteScore
6.30
自引率
5.90%
发文量
193
审稿时长
3 months
期刊介绍: PROTEOMICS is the premier international source for information on all aspects of applications and technologies, including software, in proteomics and other "omics". The journal includes but is not limited to proteomics, genomics, transcriptomics, metabolomics and lipidomics, and systems biology approaches. Papers describing novel applications of proteomics and integration of multi-omics data and approaches are especially welcome.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信