BOSO: A novel feature selection algorithm for linear regression with high-dimensional data
Luis V Valcárcel 1 2 , Edurne San José-Enériz 2 3 , Xabier Cendoya 1 , Ángel Rubio 1 4 5 , Xabier Agirre 2 3 , Felipe Prósper 2 3 6 7 , Francisco J Planes 1 4 5
With the frenetic growth of high-dimensional datasets in different biomedical domains, there is an urgent need to develop predictive methods able to deal with this complexity.
Feature selection is a relevant strategy in machine learning to address this challenge. We introduce a novel feature selection algorithm for linear regression called BOSO (Bilevel Optimization Selector Operator). We conducted a benchmark of BOSO with key algorithms in the literature, finding a superior accuracy for feature selection in high-dimensional datasets.
Proof-of-concept of BOSO for predicting drug sensitivity in cancer is presented. A detailed analysis is carried out for methotrexate, a well-studied drug targeting cancer metabolism.
CITATION PLoS Comput Biol. 2022 May 31;18(5):e1010180. doi: 10.1371/journal.pcbi.1010180. eCollection 2022 May.