DeepKbhb: Context-Aware Prediction of Human Lysine β-Hydroxybutyrylation Sites

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Accuracy: 0.856; F1-score: 0.863; Matthews Correlation Coefficient (MCC): 0.716 — all reported as superior to prior benchmark model KbhbXG

• Primary Outcome: Prediction of human lysine β-hydroxybutyrylation (Kbhb) sites from protein sequence on an independent test set

• Nnt Or Sensitivity: Sensitivity and specificity not reported in abstract; MCC of 0.716 indicates moderate-to-strong discriminative ability accounting for class imbalance, but cannot be interpreted without knowledge of dataset class distribution

• Confidence Interval: Not reported

Clinical Application

The web interface (https://awi.cuhk.edu.cn/~DeepKbhb/) provides accessible deployment for research use without requiring local computational infrastructure. Feasibility for clinical integration is premature given the absence of experimental validation and clinical outcome data. DeepKbhb has no current direct clinical application in Australian healthcare. It is not a TGA-regulated diagnostic device and is not relevant to PBS formulary decisions. For Australian researchers, it may serve as a hypothesis-generation tool in proteomics and PTM research programs funded through NHMRC or ARC. RACGP guidelines do not address computational PTM prediction tools. Potential future relevance exists in metabolic disease research (e.g., type 2 diabetes, obesity) where ketone body metabolism and Kbhb modifications are increasingly implicated, areas of active research in Australian academic medical centres. Researchers investigating human protein post-translational modifications, particularly in the context of cancer biology, metabolic disorders (including ketogenic metabolism), and immune regulation. Not directly applicable to patient care at this stage.

Abstract

Lysine β-hydroxybutyrylation (Kbhb) is a metabolism-linked post-translational modification (PTM) that plays a critical role in regulating gene expression, stress responses, and disease progression. Despite its emerging biological significance, identifying Kbhb sites remains limited due to the cost and complexity of experimental methods. Prior work such as KbhbXG is constrained by its reliance on hand-crafted features and lacks the ability to model contextual dependencies within sequences. To address this challenge, we present DeepKbhb, a deep learning framework designed for human Kbhb site identification. By integrating sequence embeddings and six engineered descriptors through a bilinear attention network, DeepKbhb effectively captures position-dependent relationships essential for accurate Kbhb site prediction. On an independent test set, DeepKbhb achieved state-of-the-art performance with an accuracy of 0.856, an F1-score of 0.863, and a Matthews correlation coefficient of 0.716. Experimental results across multiple evaluation metrics confirm the superior performance of DeepKbhb, highlighting its potential as a valuable tool for advancing Kbhb-related functional and mechanistic studies. This capability can further support disease-oriented research, particularly in cancer, metabolic disorders, and immune regulation. Further sequence analyses revealed distinct local amino acid preferences, supporting the biological relevance of our model. The web interface is accessible at https://awi.cuhk.edu.cn/~DeepKbhb/.

References

1. 1.Dong, D., Wan, J., Cai, X., Lin, Y.-C.-D., Huang, H.-Y., & Huang, H.-D. (2026). DeepKbhb: Context-aware prediction of human lysine β-hydroxybutyrylation sites. Journal of Chemical Information and Modeling. Advance online publication. https://doi.org/10.1021/acs.jcim.5c02072