BOOME: A Python package for handling misclassified disease and ultrahigh-dimensional error-prone gene expression data

In gene expression data analysis framework, ultrahigh dimensionality and measurement error are ubiquitous features. Therefore, it is crucial to correct measurement error effects and make variable selection when fitting a regression model. In this paper, we introduce a python package BOOME, which refers to BOOsting algorithm for Measurement Error in binary responses and ultrahigh-dimensional predictors. We primarily focus on logistic regression and probit models with responses, predictors, or both contaminated with measurement error. The BOOME aims to address measurement error effects, and employ boosting procedure to make variable selection and estimation.

Automated summary

This paper introduces a Python package called BOOME for addressing the issues of ultrahigh dimensionality and measurement error in gene expression data analysis. It primarily focuses on logistic regression and probit models, and aims to correct the measurement error effects and perform variable selection and estimation using boosting procedure.