Related Experiment Video
Updated: Dec 20, 2025
10:23
Author Spotlight: Optimization of Ultrashort Peptide Matrices for Colorectal Cancer Organoids
Published on: May 3, 2024
820
DeepAIPs-Pred: Predicting Anti-Inflammatory Peptides Using Local Evolutionary Transformation Images and Structural
Shahid Akbar1,2, Matee Ullah1, Ali Raza3
1Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu 610054, China.
Journal of Chemical Information and Modeling
|December 3, 2024
Summary
A new computational model, DeepAIPs-Pred, accurately predicts anti-inflammatory peptides (AIPs). This AI tool offers a faster, cost-effective method for identifying potential therapeutic peptides, aiding drug development for inflammatory diseases.
Area of Science:
- Computational biology and bioinformatics
- Drug discovery and development
- Immunology and inflammation research
Background:
- Inflammation is a critical biological process for tissue repair but chronic inflammation drives severe diseases, including autoimmune disorders.
- Anti-inflammatory peptides (AIPs) show therapeutic promise due to specificity, potency, and low toxicity.
- Traditional in vivo methods for AIP identification are costly and time-consuming, necessitating advanced computational approaches.
Purpose of the Study:
- To develop a novel computational model, DeepAIPs-Pred, for accurate prediction of anti-inflammatory peptide sequences.
- To provide a cost-effective and efficient alternative to traditional methods for identifying potential AIPs.
Main Methods:
- Utilized LBP-PSSM and LBP-SMR for evolutionary image transformation of training samples.
- Incorporated attention-based ProtBERT-BFD embedding and QLC for contextual semantic and structural feature extraction.
- Employed differential evolution (DE) for weighted feature integration, SMOTE-Tomek Links for class imbalance, and a two-layer feature selection technique.
Main Results:
- The DeepAIPs-Pred model achieved a predictive accuracy of 94.92% and an AUC of 0.97.
- Validated on two independent datasets, the model demonstrated superior performance, improving accuracy by approximately 2% and 10% over existing state-of-the-art methods.
- The self-normalized bidirectional temporal convolutional networks (SnBiTCN) were trained using optimal features for prediction.
Conclusions:
- DeepAIPs-Pred is a highly effective and reliable computational tool for predicting anti-inflammatory peptide sequences.
- The model's performance highlights its potential to significantly accelerate drug development and support research in academia.
- This AI-driven approach offers a promising avenue for identifying novel therapeutic agents for inflammatory conditions.
