Algorithms for reconstruction and analysis of metabolic networks, with an application to Neurospora crassa

• Main Author: Dreyfuss, Jonathan M.
• Language: en_US
• Published: 2016
• Subjects: Bioinformatics
• Online Access: https://hdl.handle.net/2144/15211

In this work, I have developed optimization-based algorithms to reconstruct and analyze metabolic network models, and I have applied them to the metabolism of the filamentous fungus Neurospora crassa. The developed algorithms are: (1) LInear MEtabolite Dilution Flux Balance Analysis (limed-FBA), which predicts flux while linearly accounting for metabolite dilution; (2) One-step functional Pruning (OnePrune), which removes blocked reactions with a single compact linear program; and (3) Consistent Reproduction Of growth/no-growth Phenotype (CROP), which reconciles differences between in silico and experimental gene essentiality faster than previous approaches. Together, these algorithms comprise Fast Automated Reconstruction of Metabolism (FARM). FARM was applied to reconstruct the first genome-scale model of N. crassa metabolism. This organism has played a central role in the development of twentieth-century genetics, biochemistry and molecular biology, and continues to serve as a model organism for eukaryotic biology. The N. crassa model consists of 836 metabolic genes, 257 pathways, 6 cellular compartments, and is supported by extensive manual curation of 491 literature citations. Against an independent test set of more than 300 essential/non-essential genes that were not used to train the model, it displays 93% sensitivity and specificity. The model was also used to simulate the biochemical genetics experiments originally performed on N. crassa by comprehensively predicting nutrient rescue of essential genes and synthetic lethal interactions, and providing detailed pathway-based mechanistic explanations of the predictions. The model provides a reliable computational framework for the integration and interpretation of ongoing experimental efforts in N. crassa, and the algorithms will enhance reconstruction and analysis of high-quality genome-scale metabolic models in general.