Generic placeholder image

Personalized analysis

iCAGES is designed for solving real-life problems. It uses genomic information each patient and comes up with a list of genes that are responsible for this particular patient, shedding light into personalized cancer therapy.

Generic placeholder image

Integration of prior knowledge

iCAGES prioritizes cancer driver genes for each patient, not only based on his/her genomic information, but also on prior knowledge on gene-cancer association. Such feature is realized using Phenolyzer package designed by Hui Yang in Dr. Kai Wang's lab. Using such prior knowledge generated from decades of research, iCAGES stands on the shoulder of giants and enhances its accuracy of cancer driver gene prediction.

Generic placeholder image

Machine-learning based models

iCAGES uses radial Support Vector Machine and integrates nine features measuring the deleteriousness of a mutation. Such integration greatly enhances the performance of iCAGES and can generate a more reliable output.