Vaxi-DL: An Artificial Intelligence-Enabled Platform for Vaccine Development.

Area of Science:

• Computational biology
• Bioinformatics
• Vaccinology

Background:

• Vaccine development is a lengthy and intricate process.
• Traditional methods require extensive experimental validation.
• In silico antigen screening offers a promising approach to expedite candidate identification.

Purpose of the Study:

• To present a deep learning-based method for identifying potential vaccine candidates.
• To introduce Vaxi-DL, a web-based system for accelerated vaccine discovery.
• To leverage extensive protein property data for enhanced vaccine candidate screening.

Main Methods:

• Utilized a deep learning model incorporating 18 biological and 9154 physicochemical protein properties.
• Developed the Vaxi-DL web-based system for practical application.
• Applied the technique to screen for vaccine candidates across diverse microorganisms.

Main Results:

• Successfully identified potential vaccine candidates using the deep learning approach.
• The Vaxi-DL system demonstrated efficacy in finding candidates from bacteria, protozoa, viruses, and fungi.
• The developed method significantly accelerates the initial stages of vaccine development.

Conclusions:

• Deep learning-based in silico screening is an effective strategy for rapid vaccine candidate identification.
• Vaxi-DL provides a valuable resource for researchers aiming to accelerate vaccine development.
• The integration of comprehensive protein properties enhances the accuracy of computational vaccine discovery.