Increasing consensus of context-specific metabolic models by integrating data-inferred cell functions

Area of Science:

• Systems Biology
• Metabolic Engineering
• Computational Biology

Background:

• Genome-scale metabolic models are crucial for analyzing high-throughput data and quantifying cellular functions.
• Existing methods for creating context-specific models rely on differing assumptions, leading to high variability in model content.
• This variability often exceeds differences attributed to cell types, limiting model reliability.

Purpose of the Study:

• To develop a novel framework for inferring cellular metabolic functions prior to metabolic model construction.
• To reduce variability in context-specific metabolic models generated by different algorithms.
• To improve the biological accuracy of metabolic models by incorporating data-inferred functional information.

Main Methods:

• Curated a list of essential metabolic tasks.
• Developed a framework to infer metabolic task activity from transcriptomic data.
• Applied data-inferred tasks to guide context-specific model extraction algorithms across 44 cancer cell lines.

Main Results:

• Protecting data-inferred metabolic tasks significantly decreased model variability across different extraction algorithms.
• The resulting context-specific metabolic models demonstrated improved capture of biological variability among cell lines.
• The framework enhanced consensus between models generated by diverse computational methods.

Conclusions:

• Inferring biological knowledge from omics data before model construction is key to improving consensus among metabolic modeling algorithms.
• This approach offers a pathway for developing next-generation data contextualization methods for metabolic modeling.
• The findings provide guidelines for enhancing the reliability and biological relevance of genome-scale metabolic models.