Abstract
Backgrounds: The implementation of cancer precision medicine requires signature or biomarker-based models for prognosis and therapies. Most of current efforts in this field are paying more attention to predictive accuracy, instead of molecular interpretability. Mechanism-driven strategies have only recently emerged, aiming to build models with both predictive power and explanatory power. Methods: We developed a robust gene dysregulation analysis framework with machine learning algorithms, which is capable of exploring gene dysregulations underling carcinogenesis by integrating regulatory intensity change, target expression change, and regulator's contribution to the target. We applied the framework to profile gene dysregulations in a colorectal cancer (CRC) cohort from TCGA. A prognostic model was built by choosing dysregulations related to immune process with greedy strategy. Prognostic effect of this model was validated in two independent datasets. Findings: The identified CRC-related dysregulations significantly covered known carcinogenic mechanisms and exhibited good prognostic effect. A high-accuracy prognostic model was built by utilizing immune-related gene dysregulations, and patients with better prognosis had higher abundance of immune cells. This model also has the capability of predicting therapeutic benefits for adjuvant combined chemotherapy. Interpretation: These results demonstrated that our gene dysregulation analysis framework could contribute to elucidating the mechanisms of carcinogenesis and promoting the development of explanatory models, i.e., predictive models with interpretability, for cancer precision medicine. Funding Statement: This work was supported by the grants from National Key R&D Program of China (2018YFC0910500), National Natural Science Foundation of China (81672736), Shanghai Municipal Science and Technology Major Project (2017SHZDZX01) and NIH CPTAC (Cancer Proteomic Tumor Analysis Consortium) program. Declaration of Interests: The authors declare that they have no competing interests. Ethics Approval Statement: Not required.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
