RAMClustR

RAMClustR clusters mass spectrometry features into spectra to improve grouping and annotation of non-targeted chromatographically coupled MS datasets, including integration of idMS/MS data.

Key Features:

• Unsupervised Feature Grouping: Clusters features based on mass and retention time using an unsupervised algorithm that does not assume predictable in-source phenomena.
• Integration of idMS/MS Data: Incorporates indiscriminant tandem MS (idMS/MS) data concurrently with MS data to establish relationships between features.
• Reduction of Analytical Variation: Groups multiple signals associated with a single compound into spectra to reduce quantitative analytical variation from single-feature measures.
• Decreased False Positives in Annotation: Leverages combined MS and idMS/MS information to minimize false positive annotations arising from misclassified unpredictable phenomena.
• Versatility Across Platforms: Applies to features produced by any chromatographic-spectrometric platform or feature-finding software.

Scientific Applications:

• Metabolite Identification: Facilitates identification of metabolites by clustering MS signals into spectra and integrating idMS/MS evidence.
• Data Analysis in Metabolomics: Provides a framework for feature detection and annotation in non-targeted metabolomic workflows from chromatographically coupled MS.
• Reduction of Analytical Errors: Improves accuracy of quantification and identification by reducing errors associated with relying on single features.

Methodology:

Detects features defined by mass and retention time, groups them into spectra using an unsupervised clustering algorithm independent of predictable in-source phenomena, and integrates idMS/MS data for simultaneous feature detection across MS and idMS/MS datasets.