UmetaFlow

UmetaFlow implements an automated computational workflow to preprocess, detect features, perform spectral matching, predict molecular formulas and structures, and integrate LC-MS/MS untargeted metabolomics data with GNPS molecular networking.

Key Features:

• Automated Data Processing: Integrates algorithms for data pre-processing, feature detection, spectral matching, molecular formula prediction, and structural elucidation.
• GNPS Integration: Incorporates Feature-Based Molecular Networking (FBMN) and Ion Identity Molecular Networking (IIMN) for downstream molecular networking analyses.
• Snakemake Implementation: Implemented as a Snakemake workflow to enable scalable and reproducible execution of computational tasks.
• OpenMS via pyOpenMS: Utilizes OpenMS algorithms through pyOpenMS bindings for core computational processing steps.

Scientific Applications:

• High-throughput untargeted metabolomics: Processing and analysis of large LC-MS/MS datasets for feature detection and annotation.
• Secondary metabolite discovery in actinomycetes: Analysis of complex biological samples such as actinomycetes producing secondary metabolites.
• Validation and benchmarking: Validation and benchmarking using in-house LC-MS/MS datasets and public datasets MTBLS733 and MTBLS736.

Methodology:

Computational steps include data pre-processing for quality control, feature detection and spectral matching, molecular formula and structure prediction, integration into GNPS workflows (FBMN and IIMN), implementation as a Snakemake workflow, and use of OpenMS algorithms via pyOpenMS; validation was performed using in-house LC-MS/MS data and public datasets MTBLS733 and MTBLS736.