Essential gene prediction using limited gene essentiality information–An integrative semi-supervised machine learning strategy

Abstract

Essential gene prediction helps to find minimal genes indispensable for the survival of any organism. Machine learning (ML) algorithms have been useful for the prediction of gene essentiality. However, currently available ML pipelines perform poorly for organisms with limited experimental data. The objective is the development of a new ML pipeline to help in the annotation of essential genes of less explored disease-causing organisms for which minimal experimental data is available. The proposed strategy combines unsupervised feature selection technique, dimension reduction using the Kamada-Kawai algorithm, and semi-supervised ML algorithm employing Laplacian Support Vector Machine (LapSVM) for prediction of essential and non-essential genes from genome-scale metabolic networks using very limited labeled dataset. A novel scoring technique, Semi-Supervised Model Selection Score, equivalent to area under the ROC curve (auROC), has been proposed for the selection of the best model when supervised performance metrics calculation is difficult due to lack of data. The unsupervised feature selection followed by dimension reduction helped to observe a distinct circular pattern in the clustering of essential and non-essential genes. LapSVM then created a curve that dissected this circle for the classification and prediction of essential genes with high accuracy (auROC > 0.85) even with 1% labeled data for model training. After successful validation of this ML pipeline on both Eukaryotes and Prokaryotes that show high accuracy even when the labeled dataset is very limited, this strategy is used for the prediction of essential genes of organisms with inadequate experimentally known data, such as Leishmania sp. Using a graph-based semi-supervised machine learning scheme, a novel integrative approach has been proposed for essential gene prediction that shows universality in application to both Prokaryotes and Eukaryotes with limited labeled data. The essential genes predicted using the pipeline provide an important lead for the prediction of gene essentiality and identification of novel therapeutic targets for antibiotic and vaccine development against disease-causing parasites.

PRESGENE: A web server for PRediction of ESsential GENE using integrative machine learning strategies

“…It was admissible that the previously used performance metrics e.g. Where k is the k th model among 64 total models generated with a particular parametric combination and SSMSS best (Eq 2) calculates the best scoring model (18) .…”

“…Table 1 enlists the Features and the background software packages and programming languages used for the automation of feature calculation in the PRESGENE webserver. A brief description of each of these features used for the gene essentiality prediction has been discussed in our previous work (18). Eukaryotes (Supplementary Data, Table S2).…”

“…The server hosts a total of 14 model organisms for which the ML models are trained with 1% labeled data. The prediction performance is assessed by SSMSS score as well as the semi-supervised metrics (i.e., TPR, FPR, F-measure, MCC, auROC, and accuracy) (18). For example, the outcome of the ML strategy 2 for prokaryotic model organism Escherichia coli and eukaryotic model organism Saccharomyces cerevisiae can be visualized as three circles (Figure 1C).…”

“…Here, we present a one-stop integrative web server platform PRESGENE at https://presgene.ncl.res.in for essential gene prediction in both prokaryotes and eukaryotes. PRESGENE is an online essential gene prediction server that hosts our previously published ML strategies (17,18) with a noteworthy capability of utilizing 289 biological features. This web server provides the user with two powerful ML-based prediction strategies that work accurately for essential gene prediction in the cases with ample as well as highly limited essential gene information.The user-friendly Graphical User Interface (GUI) of PRESGENE specially benefits biologists with limited knowledge of programming to implement ML-based prediction of essential genes in lesser studied organisms with limited experimental labeled data.…”

“…Towards this, we have previously developed high performance supervised(17) and semi-supervised(18) ML-based strategies for essential gene prediction with minimal gene essentiality information which have been tested on several model organisms (Supplementary data, Table S2). The supervised ML model is built for sufficient labeled imbalanced dataset whereas the semi-supervised ML model caters to the limited labeled dataset.…”