An ensemble framework for identifying essential proteins

2020

Preprint

Self Cite

Motivation: Accurately predicting essential genes using computational methods can greatly reduce the effort in finding them via wet experiments at both time and resource scales, and further accelerate the process of drug discovery. Several computational methods have been proposed for predicting essential genes in model organisms by integrating multiple biological data sources either via centrality measures or machine learning based methods. However, the methods aiming to predict human essential genes are still limited and the performance still need improve. In addition, most of the machine learning based essential gene prediction methods are lack of skills to handle the imbalanced learning issue inherent in the essential gene prediction problem, which might be one factor affecting their performance. Results:We proposed a deep learning based method, DeepHE, to predict human essential genes by integrating features derived from sequence data and protein-protein interaction (PPI) network. A deep learning based network embedding method was utilized to automatically learn features from PPI network. In addition, 89 sequence features were derived from DNA sequence and protein sequence for each gene. These two types of features were integrated to train a multilayer neural network. A cost-sensitive technique was used to address the imbalanced learning problem when training the deep neural network. The experimental results for predicting human essential genes showed that our proposed method, DeepHE, can accurately predict human gene essentiality with an average AUC higher than 94%, the area under precision-recall curve (AP) higher than 90%, and the accuracy higher than 90%. We also compared DeepHE with several widely used traditional machine learning models (SVM, Naïve Bayes, Random Forest, Adaboost). The experimental results showed that DeepHE greatly outperformed the compared machine learning models. Conclusions:We demonstrated that human essential genes can be accurately predicted by designing effective machine learning algorithm and integrating representative features captured from available biological data. The proposed deep learning framework is effective for such task.Essential genes are a subset of genes which are indispensable to the survival or reproduction of a living organism. The prediction of gene essentiality is very important for understanding the minimal requirements of an organism, identifying disease genes, and finding new drug targets. The discovery of essential genes via wet-lab experimental methods are often time-consuming, laborious, and costly. With the accumulation of gene essentiality data in some model organisms and human cell lines, many computational methods have been proposed to predict essential genes by exploring the correlations between gene essentiality and all sorts of biological information.One focus in this direction is network based centrality measures. Many studies have demonstrated that highly connected proteins in a protein-protein interaction (PPI) network are more likely to be esse...

Section: Features Derived From Sequence Datamentioning

confidence: 99%

mentioning

confidence: 99%

DeepHE: Accurately Predicting Human Essential Genes based on Deep Learning

Zhang

2020

Preprint

Self Cite

Section: Introductionmentioning

confidence: 99%

DeepHE: Accurately predicting human essential genes based on deep learning

Zhang

2020

PLoS Comput Biol

Self Cite

Accurately predicting essential genes using computational methods can greatly reduce the effort in finding them via wet experiments at both time and resource scales, and further accelerate the process of drug discovery. Several computational methods have been proposed for predicting essential genes in model organisms by integrating multiple biological data sources either via centrality measures or machine learning based methods. However, the methods aiming to predict human essential genes are still limited and the performance still need improve. In addition, most of the machine learning based essential gene prediction methods are lack of skills to handle the imbalanced learning issue inherent in the essential gene prediction problem, which might be one factor affecting their performance. We propose a deep learning based method, DeepHE, to predict human essential genes by integrating features derived from sequence data and protein-protein interaction (PPI) network. A deep learning based network embedding method is utilized to automatically learn features from PPI network. In addition, 89 sequence features were derived from DNA sequence and protein sequence for each gene. These two types of features are integrated to train a multilayer neural network. A cost-sensitive technique is used to address the imbalanced learning problem when training the deep neural network. The experimental results for predicting human essential genes show that our proposed method, DeepHE, can accurately predict human gene essentiality with an average performance of AUC higher than 94%, the area under precision-recall curve (AP) higher than 90%, and the accuracy higher than 90%. We also compare DeepHE with several widely used traditional machine learning models (SVM, Naïve Bayes, Random Forest, and Adaboost) using the same features and utilizing the same costsensitive technique to against the imbalanced learning issue. The experimental results show that DeepHE significantly outperforms the compared machine learning models. We have demonstrated that human essential genes can be accurately predicted by designing effective machine learning algorithm and integrating representative features captured from available biological data. The proposed deep learning framework is effective for such task.

“…GOS [ 14 ] integrated gene expression, orthology, subcellular localization and PINs together to predict essential proteins. Besides, Zhang et al proposed an ensemble framework that can significantly improve the prediction accuracy of traditional centrality measures by combining gene expression data and PINs [ 15 ]. In general, the integration of network topological properties and additional biological information can improve the prediction accuracy due to the increased robustness by considering multiple biological factors.…”

Section: Introductionmentioning

confidence: 99%

Predicting essential proteins by integrating orthology, gene expressions, and PPI networks

Xue

2018

PLoS ONE

Self Cite

Identifying essential proteins is very important for understanding the minimal requirements of cellular life and finding human disease genes as well as potential drug targets. Experimental methods for identifying essential proteins are often costly, time-consuming, and laborious. Many computational methods for such task have been proposed based on the topological properties of protein-protein interaction networks (PINs). However, most of these methods have limited prediction accuracy due to the noisy and incomplete natures of PINs and the fact that protein essentiality may relate to multiple biological factors. In this work, we proposed a new centrality measure, OGN, by integrating orthologous information, gene expressions, and PINs together. OGN determines a protein’s essentiality by capturing its co-clustering and co-expression properties, as well as its conservation in the evolution process. The performance of OGN was tested on the species of Saccharomyces cerevisiae. Compared with several published centrality measures, OGN achieves higher prediction accuracy in both working alone and ensemble.