TNF_ipred: a classification model to predict TNF-α inhibitors.

Area of Science:

• Biochemistry
• Immunology
• Computational Biology

Background:

• Rheumatoid arthritis (RA) is a chronic autoimmune disease causing joint inflammation.
• Tumor necrosis factor-alpha (TNF-α) overproduction significantly contributes to RA-related joint damage, pain, and swelling.
• Current TNF-α inhibitors have limitations including administration difficulties, high costs, and side effects, highlighting the need for novel small molecule inhibitors.

Purpose of the Study:

• To develop and validate a machine learning (ML) model for predicting novel TNF-α inhibitors.
• To address the limitations of traditional drug discovery methods for identifying TNF-α inhibitors.

Main Methods:

• Four classification algorithms (naïve Bayes, random forest, k-nearest neighbor, support vector machine) were trained.
• Models were trained using three feature sets: 1D, 2D, and molecular fingerprints.
• Model performance was evaluated based on accuracy and sensitivity.

Main Results:

• The random forest (RF) model demonstrated the highest performance.
• The RF model achieved an accuracy of 87.96% and a sensitivity of 86.17% when utilizing 1D, 2D, and fingerprint features.
• This represents the first ML model developed for predicting TNF-α inhibitors.

Conclusions:

• Machine learning, specifically the random forest algorithm, offers a promising and efficient approach for identifying potential TNF-α inhibitors.
• The developed ML model can accelerate the drug discovery process for rheumatoid arthritis treatments.
• The predictive model is accessible online for further research and application.