使用机器学习从MALDI-TOF质谱预测抗菌素耐药性:一个验证研究
Niklas Wiesmann1, Dominic Enders2, Antje Westendorf3
1Institute of Medical Microbiology, University of Münster, Münster, Germany.
Journal of clinical microbiology
|November 26, 2025
概括
机器学习 (ML) 模型可以使用矩阵辅助激光脱离-电离-飞行时间 (MALDI-TOF) 质谱预测抗菌素耐药性 (AMR). 定期对本地数据进行重新训练对于随着时间的推移保持模型性能至关重要.
科学领域:
- 临床微生物学和传染病.
- 计算生物学和生物信息学
- 分析化学和质谱学.
背景情况:
- 矩阵辅助激光脱离-电离-飞行时间 (MALDI-TOF) 质谱是一种快速的细菌识别方法.
- 使用MALDI-TOF数据与机器学习 (ML) 预测抗菌素耐药性 (AMR) 可以加快抗菌素敏感性测试 (AST).
- 早期预测AMR对于及时,有针对性的抗生素治疗至关重要,这可能会降低严重感染中的死亡率.
研究的目的:
- 通过使用来自常规诊断和公共数据库的MALDI-TOF数据,验证ML模型用于AMR预测的性能.
- 评估这些ML模型在18个月的长期性能和稳定性.
- 确定培训和再培训ML模型的最佳策略,以确保AMR预测的持续准确性.
主要方法:
- 收集并分析了来自大肠杆菌 (n=7,897),肺炎菌 (n=2,444) 和金黄色葡萄球菌 (n=4,664) 的MALDI-TOF质谱.
- 训练并进行交叉验证六种不同的ML分类模型 (LR,MLP,SVM,RF,LGBM,XGB) 用于AMR预测.
- 在培训后的18个月内,前性监测模型性能,评估来自本地 (德国) 和外部 (瑞士) 来源的数据.
主要成果:
- ML模型在本地和公共数据集上展示了可比性能 (AUROC ≥ 0.8),用于预测特定的AMR模式.
- 在S.中获得了对西林耐药性的最佳预测. 黄金色 (RF,0.85),西普罗夫洛克萨在E. 大肠杆菌 (XGB,0.83) 和皮佩拉林-塔扎巴克坦在K. 肺炎 (XGB,0.81) 是一个严重的疾病.
- 当训练和测试数据在地理或时间上不同时,模型的性能显著下降,这凸显了需要当地适应和定期再培训的需要.
结论:
- 基于ML的AMR预测使用MALDI-TOF光谱显示出临床诊断的重大前景,实现了良好的预测性能 (AUROC ≥0.8).
- 机器学习模型的准确性高度依赖于训练数据的位置和时间相关性.
- 用最新的,本地MALDI-TOF数据定期重新培训ML分类器对于保持AMR预测的高性能水平至关重要.
相关概念视频
MALDI-TOF Mass Spectrometry
6.4K
Mass spectrometry is a powerful characterization technique that can identify and separate a wide variety of compounds ranging from chemical to biological entities, based on their mass-to-charge ratio (m/z). The instruments that allow this detection, known as mass spectrometers, have three components: an ion source, a mass analyzer, and a detector. These spectrometers differ based on the nature of their ion source and analyzers.Matrix-assisted laser desorption ionization (MALDI) is a commonly...
6.4K
Peptide Identification Using Tandem Mass Spectrometry
8.1K
Tandem mass spectrometry, also known as MS/MS or MS2, is an analytical technique that employs two mass analyzers. Essentially it is a series of mass spectrometers that helps isolate a particular biomolecule and then helps study its chemical properties.
This technique helps gather information regarding the protein from which the peptide was obtained and to study the peptides’ amino acid sequence. Identifying peptides from a complex mixture is an important component of the growing field of...
This technique helps gather information regarding the protein from which the peptide was obtained and to study the peptides’ amino acid sequence. Identifying peptides from a complex mixture is an important component of the growing field of...
8.1K
Matrix-Assisted Laser Desorption Ionization (MALDI)
932
Matrix-assisted laser desorption ionization (MALDI) is a powerful analytical technique used in mass spectrometry. It enables the identification and characterization of various biomolecules, including proteins, peptides, nucleic acids, and carbohydrates. MALDI is an ionization technique, widely employed in biological and medical research, as well as in fields like pharmacology and biochemistry.The analyte of interest, a biomolecule or a mixture of biomolecules, is mixed with a suitable matrix...
932


