A new metabolic model of Drosophila melanogaster and the integrative analysis of Parkinson's disease

Abstract

High conservation of the disease-associated genes between flies and humans facilitates the common use of Drosophila mela-nogaster to study metabolic disorders under controlled laboratory conditions. However, metabolic modeling studies are highly lim-ited for this organism. We here report a comprehensively curated genome-scale metabolic network model of Drosophila using an orthology-based approach. The gene coverage and metabolic in-formation of the draft model derived from a reference human model were expanded via Drosophila-specific KEGG and MetaCyc databases, with several curation steps to avoid metabolic redun-dancy and stoichiometric inconsistency. Furthermore, we performed literature-based curations to improve gene-reaction associa-tions, subcellular metabolite locations, and various metabolic pathways. The performance of the resulting Drosophila model (8,230 reactions, 6,990 metabolites, and 2,388 genes), iDrosophila1 (https://github.com/SysBioGTU/iDrosophila), was assessed using flux balance analysis in comparison with the other currently available fly models leading to superior or comparable results. We also evaluated the transcriptome-based prediction capacity of iDrosophila1, where differential metabolic pathways during Parkin-son's disease could be successfully elucidated. Overall, iDrosophila1 is promising to investigate system-level metabolic alterations in re-sponse to genetic and environmental perturbations.