An integrative machine learning strategy for improved prediction of essential genes in Escherichia coli metabolism using flux-coupled features

Abstract

Prediction of essential genes helps to identify a minimal set of genes that are absolutely required for the appropriate functioning and survival of a cell. The available machine learning techniques for essential gene prediction have inherent problems, like imbalanced provision of training datasets, biased choice of the best model for a given balanced dataset, choice of a complex machine learning algorithm, and data-based automated selection of biologically relevant features for classification. Here, we propose a simple support vector machine-based learning strategy for the prediction of essential genes in Escherichia coli K-12 MG1655 metabolism that integrates a non-conventional combination of an appropriate sample balanced training set, a unique organism-specific genotype, phenotype attributes that characterize essential genes, and optimal parameters of the learning algorithm to generate the best machine learning model (the model with the highest accuracy among all the models trained for different sample training sets). For the first time, we also introduce flux-coupled metabolic subnetwork-based features for enhancing the classification performance. Our strategy proves to be superior as compared to previous SVM-based strategies in obtaining a biologically relevant classification of genes with high sensitivity and specificity. This methodology was also trained with datasets of other recent supervised classification techniques for essential gene classification and tested using reported test datasets. The testing accuracy was always high as compared to the known techniques, proving that our method outperforms known methods. Observations from our study indicate that essential genes are conserved among homologous bacterial species, demonstrate high codon usage bias, GC content and gene expression, and predominantly possess a tendency to form physiological flux modules in metabolism.

Machine and deep learning meet genome-scale metabolic modeling

“…Traditional features were shown to give low precision for the majority of gene interactions, whereas FBA-based features brought significant improvements in predictive precision and recall, indicating that genomescale CBM captures relevant information that is missed by gene-level traditional features. The approach was tested again in the context of gene essentiality prediction by Nandi and colleagues [81], who instead employed flux coupling analysis (FCA) as feature generator to take gene adaptability into account in varying environmental conditions [82].…”

The era of big data: Genome-scale modelling meets machine learning

“…In 2017, Nandi et al [123] extended the hybrid methodology utilising SVM-based implementation for binary classification of E. coli genes based on gene sequencing and expression, network topology and flux-based features. By also accounting for environmental factors, the model was able to capture the minimal set of genes that are essential in any given environment.…”

Machine Learning Methods for Analysis of Metabolic Data and Metabolic Pathway Modeling

“…Machine learning models can be trained to predict and classify genes of an organism as essential or non-essential based on a training set of known, labelled essential or non-essential genes. A number of different machine learning methods has been used to try and determine gene essentiality from metabolic network information, including: SVM, ensemble-based learning, probabilistic Bayesian methods, logistic regression and decision tree-based methods (reviewed in [ 75 ]). Plaimas et al [ 76 ] have presented a machine learning strategy for the determination of essential enzymes in a metabolic network aiming towards the determination of interesting drug targets.…”

“…An example of the ensemble learning method is random forests, which combine two machine learning techniques: bagging and random feature subset selection for predictions. Nandi et al [ 75 ] present a very interesting approach for the selection of essential features using the SVM-RFE machine learning approach. In their example, the authors used a genome-scale metabolic network of Escherichia coli to create reaction-gene combinations and label the essentiality of each combination based on experimental data.…”