A Hybrid Flux Balance Analysis and Machine Learning Pipeline Elucidates Metabolic Adaptation in Cyanobacteria.

Supreeta Vijayakumar,Claudio Angione,Pattanathu K.S.M Rahman

doi:10.1016/j.isci.2020.101818

Supreeta Vijayakumar, Claudio Angione + Show 1 more

Open Access

https://doi.org/10.1016/j.isci.2020.101818

Copy DOI

Journal: iScience	Publication Date: Nov 18, 2020
Citations: 33	License type: cc-by

Affiliation: Teesside University, University of Portsmouth

Abstract

SummaryMachine learning has recently emerged as a promising tool for inferring multi-omic relationships in biological systems. At the same time, genome-scale metabolic models (GSMMs) can be integrated with such multi-omic data to refine phenotypic predictions. In this work, we use a multi-omic machine learning pipeline to analyze a GSMM of Synechococcus sp. PCC 7002, a cyanobacterium with large potential to produce renewable biofuels. We use regularized flux balance analysis to observe flux response between conditions across photosynthesis and energy metabolism. We then incorporate principal-component analysis, k-means clustering, and LASSO regularization to reduce dimensionality and extract key cross-omic features. Our results suggest that combining metabolic modeling with machine learning elucidates mechanisms used by cyanobacteria to cope with fluctuations in light intensity and salinity that cannot be detected using transcriptomics alone. Furthermore, GSMMs introduce critical mechanistic details that improve the performance of omic-based machine learning methods.

Highlights

In the field of systems biology, several approaches have been proposed to capture the enormous complexity of biological systems by utilizing mathematical modeling and computational methods, with the goal of amalgamating the information required to build and refine predictive models
This can prove useful in refining phenotypic predictions across various environmental conditions (Vijayakumar et al, 2017; Sanchez et al, 2017; van der Ark et al, 2017; Angione, 2018) and can predict steps to engineer an organism in a way that optimizes the production of certain metabolites, which is highly applicable in many fields of industrial biotechnology including the production of biofuels, biosurfactants, and pharmaceuticals (Angione et al, 2015; Dougherty et al, 2017; Huang et al, 2017; Fatma et al, 2018; Occhipinti et al, 2018)
Aims and Objectives In this work, we present a pipeline combining metabolic modeling with statistical and machine learning tools (Figure 1) for analyzing a genome-scale metabolic models (GSMMs) of the cyanobacterium Synechococcus sp

Summary

Introduction

In the field of systems biology, several approaches have been proposed to capture the enormous complexity of biological systems by utilizing mathematical modeling and computational methods, with the goal of amalgamating the information required to build and refine predictive models. Additional constraints can be applied during FBA to shrink the solution space (Reed, 2012), providing a more accurate representation of metabolic capability as a greater number of factors can be considered to explain cellular behavior This can prove useful in refining phenotypic predictions across various environmental conditions (Vijayakumar et al, 2017; Sanchez et al, 2017; van der Ark et al, 2017; Angione, 2018) and can predict steps to engineer an organism in a way that optimizes the production of certain metabolites, which is highly applicable in many fields of industrial biotechnology including the production of biofuels, biosurfactants, and pharmaceuticals (Angione et al, 2015; Dougherty et al, 2017; Huang et al, 2017; Fatma et al, 2018; Occhipinti et al, 2018)

Objectives

Methods

Results

Discussion

Conclusion