McKetney Justin, Miller Ian J, Hutton Alexandre, Sinitcyn Pavel, Serrano Lia R, Coon Joshua J, Meyer Jesse G
Department of Biomolecular Chemistry, University of Wisconsin-Madison, Madison, Wisconsin 53706, United States.
National Center for Quantitative Biology of Complex Systems, Madison, Wisconsin 53706, United States.
Anal Chem. 2025 Feb 4;97(4):2254-2263. doi: 10.1021/acs.analchem.4c05359. Epub 2025 Jan 26.
Peptide ion mobility adds an extra dimension of separation to mass spectrometry-based proteomics. The ability to accurately predict peptide ion mobility would be useful to expedite assay development and to discriminate true answers in a database search. There are methods to accurately predict peptide ion mobility through drift tube devices, but methods to predict mobility through high-field asymmetric waveform ion mobility (FAIMS) are underexplored. Here, we successfully model peptide ions' FAIMS mobility using a multi-label classification scheme to account for non-normal transmission distributions. We trained two models from over 100,000 human peptide precursors: a random forest and a long-term short-term memory (LSTM) neural network. Both models had different strengths, and the ensemble average of model predictions produced a higher F2 score than either model alone. Finally, we explored cases where the models make mistakes and demonstrate the predictive performance of F2 = 0.66 (AUROC = 0.928) on a new test data set of nearly 40,000 peptide ions. The deep learning model is easily accessible via https://faims.xods.org.
肽离子淌度为基于质谱的蛋白质组学增加了一个额外的分离维度。准确预测肽离子淌度的能力将有助于加快分析方法的开发，并在数据库搜索中辨别正确答案。有一些方法可以通过漂移管装置准确预测肽离子淌度，但通过高场不对称波形离子淌度（FAIMS）预测淌度的方法尚未得到充分探索。在这里，我们使用多标签分类方案成功地对肽离子的FAIMS淌度进行建模，以考虑非正态传输分布。我们从超过100,000个人类肽前体中训练了两个模型：一个随机森林模型和一个长短期记忆（LSTM）神经网络。两个模型各有优势，模型预测的总体平均值产生的F2分数高于单独的任何一个模型。最后，我们探讨了模型出错的情况，并在一个近40,000个肽离子的新测试数据集上展示了F2 = 0.66（AUROC = 0.928）的预测性能。深度学习模型可通过https://faims.xods.org轻松访问。
