In bioinformatics, the rapid development of sequencing technology has enabled us to collect an increasing amount of omics data. Classification based on omics data is one of the central problems in biomedical research. However, omics data usually has a limited sample size but high feature dimensions, and it is assumed that only a few features (biomarkers) are active, i.e. informative to discriminate between different categories. Identifying active biomarkers for classification has therefore become fundamental for omics data analysis. Focusing on binary classification, we propose an innovative feature selection method aiming at dealing with the high correlations between the biomarkers. Our method, WLogit, consists in whitening the design matrix to remove the correlations between biomarkers, then using a penalized criterion adapted to the logistic regression model to select features. The results from numerical experiments suggest that WLogit can identify almost all active biomarkers even in the cases where the biomarkers are highly correlated, while the other methods fail, which consequently leads to higher classification accuracy. The performance of WLogit is also evaluated on two publicly available datasets, and the obtained classifier outperformed other methods in terms of prediction accuracy. Our method is implemented in the WLogit R package available from the Comprehensive R Archive Network (CRAN).

In clinical trials, identification of prognostic and predictive biomarkers has became essential to precision medicine. Prognostic biomarkers can be useful for the prevention of the occurrence of the disease, and predictive biomarkers can be used to identify patients with potential benefit from the treatment. Previous researches were mainly focused on clinical characteristics, and the use of genomic data in such an area is hardly studied. A new method is required to simultaneously select prognostic and predictive biomarkers in high dimensional genomic data where biomarkers are highly correlated. We propose a novel approach called PPLasso, that integrates prognostic and predictive effects into one statistical model. PPLasso also takes into account the correlations between biomarkers that can alter the biomarker selection accuracy. Our method consists in transforming the design matrix to remove the correlations between the biomarkers before applying the generalized Lasso. In a comprehensive numerical evaluation, we show that PPLasso outperforms the traditional Lasso and other extensions on both prognostic and predictive biomarker identification in various scenarios. Finally, our method is applied to publicly available transcriptomic and proteomic data.

MOTIVATION: In genomic studies, identifying biomarkers associated with a variable of interest is a major concern in biomedical research. Regularized approaches are classically used to perform variable selection in high-dimensional linear models. However, these methods can fail in highly correlated settings. RESULTS: We propose a novel variable selection approach called WLasso, taking these correlations into account. It consists in rewriting the initial high-dimensional linear model to remove the correlation between the biomarkers (predictors) and in applying the generalized Lasso criterion. The performance of WLasso is assessed using synthetic data in several scenarios and compared with recent alternative approaches. The results show that when the biomarkers are highly correlated, WLasso outperforms the other approaches in sparse high-dimensional frameworks. The method is also illustrated on publicly available gene expression data in breast cancer. AVAILABILITYAND IMPLEMENTATION: Our method is implemented in the WLasso R package which is available from the Comprehensive R Archive Network (CRAN). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Summary: The R package biospear allows selecting the biomarkers with the strongest impact on survival and on the treatment effect in high-dimensional Cox models, and estimating expected survival probabilities. Most of the implemented approaches are based on penalized regression techniques. Availability and implementation: The package is available on the CRAN. (https://CRAN.R-project.org/package=biospear). Contact: [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.