MEANtools integrates multi-omics data to identify metabolites and predict biosynthetic pathways.

Area of Science:

• Plant biochemistry and specialized metabolism
• Computational biology and bioinformatics
• Metabolomics and transcriptomics

Background:

• Plants produce diverse specialized metabolites crucial for adaptation, but their biosynthetic pathways are often unknown.
• Current methods for pathway elucidation rely on target-based approaches requiring prior knowledge.
• Integrating genomics, transcriptomics, and metabolomics offers potential but faces challenges.

Purpose of the Study:

• To present MEANtools, a novel, unsupervised computational workflow for predicting plant metabolic pathways de novo.
• To leverage general reaction rules and metabolic structures for pathway discovery.
• To enable hypothesis generation for novel biosynthetic pathways.

Main Methods:

• MEANtools uses a systematic, unsupervised computational integrative omics workflow.
• It identifies connections between metabolites and transcripts with correlated abundance using reaction rules.
• The workflow assesses reactions linking transcript-correlated mass features within candidate pathways.

Main Results:

• MEANtools successfully predicted five out of seven steps in the tomato falcarindiol biosynthetic pathway.
• The tool identified other candidate pathways involved in specialized plant metabolism.
• Validation demonstrated MEANtools' potential for hypothesis generation in pathway discovery.

Conclusions:

• MEANtools represents a significant advancement in integrating multi-omics data for biochemical pathway elucidation.
• The unsupervised approach overcomes limitations of target-based methods.
• This tool has broad applications for understanding plant metabolism and beyond.