Model-driven analysis of mutant fitness experiments improves genome-scale metabolic models of Zymomonas mobilis ZM4

摘要

We have used several methods of analysis of a high-throughput transposon library dataset for the organism Zymomonas mobilis in order to improve a newly developed genome-scale metabolic model. Data from individual experiments within this dataset were compared to predictions made by our draft model to determine areas of inaccuracies in the model. Correlations within the dataset were investigated in the framework of metabolic modules in the model enabling us to identify, and subsequently experimentally confirm, genes for several reactions in the model that previously lacked an associated gene. Finally, we looked at which metabolic modules correlate poorly, highlighting where metabolic knowledge gaps in Z. mobilis may still reside. Our metabolic model, iZM4_478, is one of the most complete for Z. mobilis ZM4 to date, and we expect it will be a valuable tool for investigating the unique, streamlined metabolism of this organism. The data analysis approaches that we used can be easily applied by other investigators in the development of metabolic models for other organisms with similar datasets, leading to more complete assignment of genes to reactions and more accurate metabolic models.