Mining heterogeneous causal effects for personalized cancer treatment

Personalized Medicine
DOI: 10.1093/bioinformatics/btx174 Publication Date: 2017-03-24T03:10:34Z
ABSTRACT
Abstract Motivation Cancer is not a single disease and involves different subtypes characterized by sets of molecules. Patients with cancer often react heterogeneously towards the same treatment. Currently, clinical diagnoses rather than molecular profiles are used to determine most suitable A level approach will allow more precise informed way for making treatment decisions, leading better survival chance less suffering patients. Although many computational methods have been proposed identify at level, best our knowledge none them designed discover heterogeneous responses. Results In this article we propose Survival Causal Tree (SCT) method. SCT patient subgroups effects from censored observational data. on TCGA breast invasive carcinoma glioma datasets shown that each subtype identified SCT, patients treated radiotherapy exhibit significantly relapse free pattern when compared without With capability responses, useful in helping choose individual Availability Implementation Data code available https://github.com/WeijiaZhang24/SurvivalCausalTree. Supplementary information data Bioinformatics online.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (40)
CITATIONS (29)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....