RESUMO
Machine learning (ML) has the potential to identify subsets of patients with distinct phenotypes from gene expression data. However, phenotype prediction using ML has often relied on identifying important genes without a systems biology context. To address this, we created an interpretable ML approach based on blood transcriptomics to predict phenotype in systemic lupus erythematosus (SLE), a heterogeneous autoimmune disease. We employed a sequential grouped feature importance algorithm to assess the performance of gene sets, including immune and metabolic pathways and cell types, known to be abnormal in SLE in predicting disease activity and organ involvement. Gene sets related to interferon, tumor necrosis factor, the mitoribosome, and T cell activation were the best predictors of phenotype with excellent performance. These results suggest potential relationships between the molecular pathways identified in each model and manifestations of SLE. This ML approach to phenotype prediction can be applied to other diseases and tissues.
RESUMO
Because of the clinical heterogeneity among patients with systemic lupus erythematosus (SLE), developing molecular profiles that predict clinical features can be useful in creating a personalized approach to treatment. Toro-Domínguez et al. created a web tool to aid in therapeutic decision making for clinicians that predicts clinical features associated with SLE from blood transcriptomic data. Specifically, they present a machine learning model that predicts the presence of proliferative nephritis from blood transcriptomics. Here, we report use of the tool in independent datasets and found that it did not perform sufficiently well to consider replacement of the standard kidney biopsy as a diagnostic procedure.
