GrowMatch: An Automated Method for Reconciling In Silico/In Vivo Growth Predictions
Author Information
Author(s): Kumar Vinay Satish, Maranas Costas D.
Primary Institution: The Pennsylvania State University
Hypothesis
Can an optimization-based framework automatically reconcile inconsistencies between in silico growth predictions and in vivo data?
Conclusion
GrowMatch improved the consistency of in silico growth predictions from 90.6% to 96.7% by resolving inconsistencies in metabolic models.
Supporting Evidence
- GrowMatch suggested consistency-restoring hypotheses for 56 out of 72 GNG mutants.
- GrowMatch resolved 18 GNG inconsistencies by suggesting suppressions in the metabolic networks.
- GrowMatch improved the specificity of growth predictions from 67.6% to 79.3% with global corrections.
Takeaway
The study created a tool called GrowMatch that helps scientists fix mistakes in computer models of bacteria by comparing them to real-life experiments.
Methodology
An optimization-based framework was used to reconcile growth prediction inconsistencies by suggesting model modifications.
Limitations
The method relies on existing data and may miss biologically relevant hypotheses that require more than the allowed modifications.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website