Editorial: Application of bioinformatics, machine learning, and artificial intelligence to improve diagnosis, prognosis and treatment of cancer

Editorial on the Research Topic Application of bioinformatics, machine learning, and artificial intelligence to improve diagnosis, prognosis and treatment of cancerIn recent years, omics approaches have yielded great advances in cancer research and have provided new in-depth insights into the processes involved in cancer development and progression.Practical use of the information contained within this huge amount of data requires computational approaches such as bioinformatics, machine learning (ML), and artificial intelligence (AI).These computational methods, together with omics data from large databases, such as The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO), can now be used to develop cancer biomarkers, novel anti-cancer drug targets, and both novel and repurposed treatment options for cancer.Considering the application of versatile computational methods in cancer research, we collected original research articles in this Research Topic to present the novel discovery of potential cancer drug targets, prognostic biomarkers, or therapeutic interventions.Lower extremity deep vein thrombosis (DVT) is a frequent postoperative complication, occurring in up to 40% of patients with colorectal cancer.Liu et al. used an ML model optimized for predicting an individual's risk of DVT in colorectal cancer patients.Given the prevalence of DVT and that traditional risk assessments may not be accurate indicators of true risk, they showed that the XGBoost model (Chen and Guestrin, 2016) has strong potential for improving early detection and treatment in clinical settings.Identifying novel biomarkers for predicting patient survival time is of crucial practical clinical significance, since it could lead to better patient stratification and treatment decisions.Li et al. used the reactive oxygen species (ROS)-related signature genes, which they identified using the TCGA data, to predict the prognosis and chemotherapy response of patients with bladder cancer.They did not only identify 17 ROS-related genes that exhibited good overall survival in bladder cancer patients, but also 11 potential small