Cancer drug sensitivity prediction method and system, storage medium and terminal

Abstract

The invention provides a cancer drug sensitivity prediction method and system, a storage medium and a terminal. The cancer drug sensitivity prediction method comprises the following steps: acquiring proteome and phosphorylated proteome data and drug sensitivity values of a plurality of patients; obtaining edge strength characteristics or node characteristic values of proteome and phosphorylated proteome data, and dividing the multiple drug effect sensitive values and the corresponding edge strength characteristics or node characteristic values into a training data set and a test data set; training an elastic network based on the training data set to predict cancer drug sensitivity, the elastic network being validated based on the test data set; and predicting the cancer drug sensitivity based on the trained elastic network. According to the cancer drug sensitivity prediction method and system, the storage medium and the terminal, drug effect discrimination of the targeted drug is realized by adopting the machine learning model based on the kinase substrate feature edge, so that accurate prediction of the cancer drug sensitivity is provided.

Description

technical field [0001] The present invention relates to the technical field of cancer drug sensitivity prediction, in particular to a cancer drug sensitivity prediction method, system, storage medium and terminal. Background technique [0002] Colorectal cancer is the fourth most deadly cancer in the world, and about 900,000 people die from colorectal cancer every year. Aging of the population, unhealthy modern diet and other variable environmental factors (including smoking, obesity and low physical activity) are all risk factors that increase the prevalence of colorectal cancer. Colorectal cancer usually arises from polyps in the lining of the colon or rectum, and it usually takes 10-20 years to develop from polyps to cancer. More than 96 percent of colon cancers are of this type, also known as adenocarcinoma, that grow from the lining. Once a tumor in the intestine breaks through the wall of the colon or rectum, it may spread into the blood or lymph nodes, and the cance...

AI Technical Summary

Problems solved by technology

There is not only intra-tumor heterogeneity, but also huge heterogeneity among metastatic tissues and different individuals, which also brings great challenges to the research of colorectal cancer, especially metastatic colorectal cancer

Images