AI-driven genotype-phenotype modeling: a framework integrating multi-modal single-cell genomics and reverse vaccinology for de novo design of multi-epitope cancer vaccines

Description

Mahabub Mallik and Sahid Afrid Mollick wrote a 2026 review proposing a framework for designing personalized cancer vaccines with artificial intelligence. Bulk sequencing averages millions of cells and hides the variety inside a tumor, which drives resistance and immune escape. Single-cell methods now map that variety in fine detail. Their framework filters vaccine targets in two stages: single-cell data first weeds out false targets and clones that have lost the target.

The cleaned data then feeds structural and generative AI models that design vaccines with several targets at once. The review surveys the deep learning tools for each step, from spotting tumor clones to predicting how vaccine pieces bind immune molecules. It closes with the hard parts: models that are hard to interpret, shifting tumors, and factory limits.

Catalog record

Source: https://europepmc.org/articles/PMC13612945