Meta-Analyzing Multiple Omics Data With Robust Variable Selection

Abstract

High-throughput omics data are becoming more and more popular in various areas of science. Given that many publicly available datasets address the same questions, researchers have applied meta-analysis to synthesize multiple datasets to achieve more reliable results for model estimation and prediction. Due to the high dimensionality of omics data, it is also desirable to incorporate variable selection into meta-analysis. Existing meta-analyzing variable selection methods are often sensitive to the presence of outliers, and may lead to missed detections of relevant covariates, especially for lasso-type penalties. In this paper, we develop a robust variable selection algorithm for meta-analyzing high-dimensional datasets based on logistic regression. We first search an outlier-free subset from each dataset by borrowing information across the datasets with repeatedly use of the least trimmed squared estimates for the logistic model and together with a hierarchical bi-level variable selection technique. We then refine a reweighting step to further improve the efficiency after obtaining a reliable non-outlier subset. Simulation studies and real data analysis show that our new method can provide more reliable results than the existing meta-analysis methods in the presence of outliers.

Springer: An R package for bi-level variable selection of high-dimensional longitudinal data

“…Robust penalization methods have drawn increasing attention in recent years ( Freue et al, 2019 ; Hu et al, 2021 ; Chen et al, 2022 ; Sun et al, 2022 ). In high-dimensional longitudinal studies, incorporation of robustness is more challenging.…”

Detecting predicted cancer-testis antigens in proteomics datasets of healthy and tumoral samples

“…The availability of raw and processed MS data has be-1 come routine for proteomics studies in recent years, and the construction of a consensus for MS data sharing allowed for the evaluation, reuse and comparative analysis of such data [15] and the consequent integration of information through meta-analysis techniques [16]. Meta-analysis is an efficient strategy for the integration and analysis of datasets from multiple studies and may provide more significant and reliable results [17], [18]. For example, the comparison of independently collected MS datasets allowed the integration of pancreatic cancer proteomic data to uncover proteins relevant to diagnosis and prognosis of pancreatic ductal adenocarcinoma [19], where the authors identified 39 secreted proteins with potential for serving as biomarkers.…”

A Novel Meta-Analysis-Based Regularized Orthogonal Matching Pursuit Algorithm to Predict Lung Cancer with Selected Biomarkers

“…There are three main types of meta-analysis methods [16]: meta-analysis based on combining results from different studies (e.g., effect sizes, p values or ranks), metaanalysis based on particular cross-platform normalization and a unified model on multiple datasets without data merging that can account for the joint effects of genes on clinical outcomes. Considering the joint modeling of multiple genes, Li et al [17] proposed metalasso, and then meta-nonconvex methods emerged gradually for solving the heterogeneity problem [18,19]. Thus, we combine the MS-ROMP strategy with meta-analysis techniques to improve the strength across multiple datasets.…”