automRm

automRm performs automatic preprocessing of liquid chromatography–triple quadrupole mass spectrometry (LC-QQQ-MS) multiple reaction monitoring (MRM) data to detect, integrate, and quality-control chromatographic peaks using machine learning for consistent peak recognition.

Key Features:

• Machine Learning Integration: Employs machine learning algorithms to detect chromatographic peaks and assess peak quality from raw LC-QQQ-MS MRM data, including recognition of complex patterns.
• Fully Automatic Preprocessing: Automates peak detection, peak integration, and quality control for targeted LC-MS (MRM) workflows, reducing reliance on manual peak review.
• Sensitive Peak Detection: Improves detection of low-abundance chromatographic peaks in targeted LC-QQQ-MS datasets.
• Applicability to Analytical Methods: Handles sample-to-sample variation and is applicable to datasets from analytical methods such as hydrophilic interaction liquid chromatography (HILIC)-MS.
• Enhanced Accuracy and Reproducibility: Reduces variation in peak integration among replicate measurements and can reproduce results comparable to manual review.
• Training and Algorithm Flexibility: Performance is influenced by choice of training datasets, selection of ML algorithms, and individual peak characteristics, allowing optimization for specific experimental conditions.

Scientific Applications:

• Targeted LC-MS Preprocessing: Preprocesses MRM-mode LC-QQQ-MS data to provide consistent peak tables for downstream analyses.
• Bioinformatics Research: Supports bioinformatics studies requiring automated, reproducible peak detection and integration.
• Analytical Chemistry: Assists analytical chemistry workflows in handling variations in retention time and peak shape across samples.
• High-Throughput Studies: Enables high-throughput targeted LC-MS experiments where manual peak review is impractical and consistent preprocessing is required for downstream statistical and biological interpretation.

Methodology:

Uses machine learning algorithms to detect chromatographic peaks, perform peak quality assessment and integration on MRM-mode LC-QQQ-MS data, with performance dependent on training data, algorithm choice, and peak characteristics.