Predicting the multi-label protein subcellular localization through multi-information fusion and MLSI dimensionality reduction based on MLFE classifier

Abstract

Motivation Multi-label protein subcellular localization (SCL) is an indispensable way to study protein function. It can locate a certain protein (such as the human transmembrane protein that promotes the invasion of the SARS-CoV-2) or expression product at a specific location in a cell, which can provide a reference for clinical treatment of diseases such as COVID-19. Results The paper proposes a novel method named ML-locMLFE. First of all, six feature extraction methods are adopted to obtain protein effective information. These methods include pseudo amino acid composition (PseAAC), encoding based on grouped weight (EBGW), gene ontology (GO), multi-scale continuous and discontinuous (MCD), residue probing transformation (RPT) and evolutionary distance transformation (EDT). In the next part, we utilize the multi-label information latent semantic index (MLSI) method to avoid the interference of redundant information. In the end, multi-label learning with feature induced labeling information enrichment (MLFE) is adopted to predict the multi-label protein SCL. The Gram-positive bacteria dataset is chosen as a training set, while the Gram-negative bacteria dataset, virus dataset, newPlant dataset and SARS-CoV-2 dataset as the test sets. The overall actual accuracy (OAA) of the first four datasets is 99.23%, 93.82%, 93.24%, and 96.72% by the leave-one-out cross validation (LOOCV). It is worth mentioning that the OAA prediction result of our predictor on the SARS-CoV-2 dataset is 72.73%. The results indicate that the ML-locMLFE method has obvious advantages in predicting the SCL of multi-label protein, which provides new ideas for further research on the SCL of multi-label protein. Availability and implementation The source codes and data are publicly available at https://github.com/QUST-AIBBDRC/ML-locMLFE/. Supplementary information Supplementary data are available at Bioinformatics online.

A Review for Artificial Intelligence Based Protein Subcellular Localization

“…But knowledge data are limited and only applicable to wellcurated proteins, which limits the predictive power of this kind of method for novel or newly discovered proteins. In recent studies [75][76][77], different kinds of information are fused together for better model performance, given that computational methods excel with high dimensional data as inputs.…”

“…The fusion methods can basically be divided into two categories: feature-level fusion [77,104,105] and decision-level fusion [106]. Feature-level fusion is mostly based on average pooling, weighted combination [107], serial combination, or concatenation of selected values.…”

“…Feature-level fusion is mostly based on average pooling, weighted combination [107], serial combination, or concatenation of selected values. Liu et al [77] utilized the latent semantic index method to represent multi-label information, while Yu et al [49] constructed a novel parallel framework of attribute fusion to avoid the impact of duplicated information. This fusion level enhances the information from multiple sources and allows flexibility in fusion techniques, such as early integration, intermediate integration, and late integration [108].…”

A Review for Artificial Intelligence Based Protein Subcellular Localization

“…But knowledge data are limited and only applicable to wellcurated proteins, which limits the predictive power of this kind of method for novel or newly discovered proteins. In recent studies [75][76][77], different kinds of information are fused together for better model performance, given that computational methods excel with high dimensional data as inputs.…”

Recent Advances in the Prediction of Subcellular Localization of Proteins and Related Topics

“…Most of them use general sequence features rather than hand-crafted features related to specific sorting signals and claim to be able to address the problem of proteins localized at multiple sites (i.e., multi-labeled proteins), though there still remains the problem that their training data do not seem to have been annotated with a uniform criterion (see below). Some of them proposed extensions of existing sequence features, such as the k -mer compositions ( Li et al, 2019 ; Yao et al, 2019 ; Sahu et al, 2020 ), while some imported external information, such as Gene Ontology and protein-protein interactions ( Chen et al, 2021 ; Liu et al, 2021 ; Zhang et al, 2021 ). One method employed an ensemble approach of multiple classifiers with voting, claiming that the approach is effective in addressing the problem of imbalanced sizes of training data between different localization sites ( Wattanapornprom et al, 2021 ).…”