ROSes-FINDER: a multi-task deep learning framework for accurate prediction of microorganism reactive oxygen species scavenging enzymes

Abstract

Reactive oxygen species (ROS) are highly reactive molecules that play important roles in microbial biological processes. However, excessive accumulation of ROS can lead to oxidative stress and cellular damage. Microorganism have evolved a diverse suite of enzymes to mitigate the harmful effects of ROS. Accurate prediction of ROS scavenging enzymes classes (ROSes) is crucial for understanding the mechanisms of oxidative stress and developing strategies to combat related diseases. Nevertheless, the existing approaches for categorizing ROS-related proteins exhibit certain drawbacks with regards to their precision and inclusiveness. To address this, we propose a new multi-task deep learning framework called ROSes-FINDER. This framework integrates three component methods using a voting-based approach to predict multiple ROSes properties simultaneously. It can identify whether a given protein sequence is a ROSes and determine its type. The three component methods used in the framework are ROSes-CNN, which extracts raw sequence encoding features, ROSes-NN, which predicts protein functions based on sequence information, and ROSes-XGBoost, which performs functional classification using ensemble machine learning. Comprehensive experiments demonstrate the superior performance and robustness of our method. ROSes-FINDER is freely available at https://github.com/alienn233/ROSes-Finder for predicting ROSes classes.

Hierarchical multi-task deep learning-assisted construction of human gut microbiota reactive oxygen species-scavenging enzymes database

“…This deep learning approach has outperformed traditional best-hit methods in terms of accuracy, effectively addressing the limitations imposed by stringent cutoff values and significantly lowering the incidence of false negatives. Our prior research ( 12 ) reveals that the ROSes-FINDER achieves a false-positive rate of 1.17%, a false-negative rate of 5.68%, and an overall accuracy rate of 94.29%. The success of ROSes-FINDER is attributed to its deep learning-driven predictive capability.…”

“…This framework employs a hierarchical prediction strategy, deploying a tiered structure for ROSes scavenging enzyme classification. Given a protein sequence, ROSes-FINDER ( 12 ) ( https://github.com/alienn233/ROSes-Finder ) first classifies it as a ROSes or non-ROSes. If the input sequence is a ROSes, we predict which ROSes category it belongs to.…”

“…Gene prediction and basic annotation were performed to identify ROSes candidates. The mining module integrates our previous ROSes-FINDER algorithm ( 12 ), employing machine learning techniques (xgboost), convolutional neural networks (CNNs), and deep neural networks, along with a voting-based approach. This combination, incorporating protein annotation, functional analysis, and sequence characteristics, allows for the reannotation and exploration of ROSes entries.…”