DeepmRNALoc: A Novel Predictor of Eukaryotic mRNA Subcellular Localization Based on Deep Learning

Abstract

The subcellular localization of messenger RNA (mRNA) precisely controls where protein products are synthesized and where they function. However, obtaining an mRNA's subcellular localization through wet-lab experiments is time-consuming and expensive, and many existing mRNA subcellular localization prediction algorithms need to be improved. In this study, a deep neural network-based eukaryotic mRNA subcellular location prediction method, DeepmRNALoc, was proposed, utilizing a two-stage feature extraction strategy that featured bimodal information splitting and fusing for the first stage and a VGGNet-like CNN module for the second stage. The five-fold cross-validation accuracies of DeepmRNALoc in the cytoplasm, endoplasmic reticulum, extracellular region, mitochondria, and nucleus were 0.895, 0.594, 0.308, 0.944, and 0.865, respectively, demonstrating that it outperforms existing models and techniques.

Keywords: artificial intelligence; chaos-game representation; deep learning; mRNA subcellular localization.

Figure 1

Performance comparison of DeepmRNALoc, SubLocEP, mRNALoc, iLoc-mRNA, and RNATracker under different evaluation indicators.

Figure 2

Accuracy with various k values.

Figure 3

Screenshots of the DeepmRNALoc web server. (A) The web interface; (B) input data upload and information-filling interface; and (C) result download page.

Figure 4

An illustration of DeepmRNALoc architecture. The CNN module contains four CNN blocks, with each CNN block consisting of two CNN layers. The BiLSTM module consists of two BiLSTM layers. The FCN module consists of four fully connected neural network layers.