iAnt: Combination of Convolutional Neural Network and Random Forest Models Using PSSM and BERT Features to Identify Antioxidant Proteins

Abstract

Background: Reactive oxygen species (ROS) has many roles in the body such as cell signaling, homeostasis or protection from harmful bacteria. However, too much ROS in the body will damage lipids, proteins, and DNA. Many studies show that many environmental factors increase the amount of ROS produced in the body. Antioxidant proteins are responsible for neutralizing these ROS or free radicals. Although the amount of data on protein sequences has increased over the last two decades, we still lack bioinformatics tools to be able to accurately identify antioxidant protein sequences while biochemical methods to determine antioxidant proteins are very expensive and time consuming, so a machine learning approach must be used to speed up the computation. In this study. Methods: we propose a new method that combines convolutional neural network and Random Forest using two features, the normalized PSSM and the best selected feature of the ProtBert output. Result: Our model gave very good results on the independent test dataset with 97.3% sensitivity and 95.9% specificity. Comparison with current state of the art models shows that our model is superior. Conclusion: We have also installed iAnt as an online web site with a friendly interface available at http://antixiodant.nguyenhongquang.edu.vn. iAnt has been developed to accurately identify the antioxidant protein. It shows results outperforming the existing state-of-the-art methods, and it is available online.

ACP-GBDT: An improved anticancer peptide identification method with gradient boosting decision tree

“…Compared with the GBDT algorithm, XGBoost maximizes speed and efficiency. Random forest (RF) is an effective machine learning algorithm ( Ao et al, 2022b ; Tran and Nguyen, 2022 ; Naik et al, 2023 ) which is a random composition of many unrelated decision trees. When judging the category of a new sample, each RF decision tree makes an independent judgment and finally selects the category with the highest probability value.…”

“…Compared with the GBDT algorithm, XGBoost maximizes speed and efficiency. Random forest (RF) is an effective machine learning algorithm (Ao et al, 2022b;Tran and Nguyen, 2022;Naik et al, 2023) which is a random composition of many Finally, a strong classifier will be obtained when the minimum error rate or the maximum number of iterations is reached. The decision tree classification algorithm constructs a tree-type classification model from the training samples (Shabbir et al, 2021).…”

Prediction of apoptosis protein subcellular location based on amphiphilic pseudo amino acid composition

“…We need to convert sequences into vectors in mathematical representation (Amanatidou, and Dedoussis, 2021;Dao et al, 2022a;Jeon et al, 2022;Li H et al, 2022;Nidhi et al, 2022;Sun et al, 2022;Tran and Nguyen, 2022;Wang et al, 2022;Yang et al, 2022;. The amino acid composition (ACC) of the protein has a great impact on its subcellular location (Chou and Elrod, 1999a;Awais et al, 2021;Chou and Elrod, 1999b;Rout et al, 2022;Naseer et al, 2021;Manavalan and Patra, 2022;Shoombuatong et al, 2022).…”

Machine learning‐based antioxidant protein identification model: Progress and evaluation

“…We have compiled relevant research conducted by researchers in recent years and compared 11 models. In the process of feature extraction, there are many aspects of feature extraction, such as amino acid composition, protein secondary structure information, and physical and chemical properties of protein sequences, which play an important role in the identification of antioxidant proteins 5–7 . In the process of feature selection, we should not only select feature combinations with high contribution but also consider that the dimension of features should not be too high 8–10 .…”

“…In the process of feature extraction, there are many aspects of feature extraction, such as amino acid composition, protein secondary structure information, and physical and chemical properties of protein sequences, which play an important role in the identification of antioxidant proteins. [5][6][7] In the process of feature selection, we should not only select feature combinations with high contribution but also consider that the dimension of features should not be too high. [8][9][10] A dimension that is too high will affect not only the efficiency of the model but also the accuracy of the model due to redundant features.…”