A supervised machine learning-based methodology for analyzing dysregulation in splicing machinery: An application in cancer diagnosis
作者：
Highlights：
• A novel supervised methodology for analyzing dysregulation in splicing machinery.
• Allows the determination of rankings of splicing components.
• Explains individual predictions and global behaviors of induced classifiers.
• A heuristic search finds accurate classification models and small subsets of relevant splicing components.
• Better exploitation and acceptance of black-box classifiers in alternative splicing studies.
摘要
•A novel supervised methodology for analyzing dysregulation in splicing machinery.•Allows the determination of rankings of splicing components.•Explains individual predictions and global behaviors of induced classifiers.•A heuristic search finds accurate classification models and small subsets of relevant splicing components.•Better exploitation and acceptance of black-box classifiers in alternative splicing studies.
论文关键词：Transcript-based analysis,Alternative Splicing,Feature weighting methods,Classification methods,Explaining classifier’s predictions
论文评审过程：Received 15 November 2019, Revised 15 August 2020, Accepted 18 August 2020, Available online 20 August 2020, Version of Record 1 September 2020.
论文官网地址：https://doi.org/10.1016/j.artmed.2020.101950
