Journal: Research Square

Article Title: Stabl: sparse and reliable biomarker discovery in predictive modeling of high-dimensional omic data

doi: 10.21203/rs.3.rs-2609859/v1

Figure Lengend Snippet: a . Clinical case study 1: Classification of individuals with normotensive pregnancy or preeclampsia (PE) from the analysis of circulating cell-free RNA (cfRNA) sequencing data. Number of samples (n) and features (p) are indicated. b. UMAP visualization of the cfRNA transcriptomic features, node size and color are proportional to the strength of the association with the outcome calculated as the p-value in a univariate Mann-Whitney test using a −log10 scale. c . Clinical case study 2: Classification of mild vs. severe COVID-19 in two independent patient cohorts from the analysis of plasma proteomic data (Olink). d. UMAP visualization of the proteomic data. Node characteristics as in ( b ). e. Predictivity performances of Stabl and Lasso for the PE datasets. AUROC Stabl = 0.83 [0.76, 0.90] , AUROC Lasso = 0.84 [0.78, 0.90] (p-value = 0.28, Bootstrap test); AUPRC Stabl = 0.85 [0.77, 0.93] , AUPRC Lasso = 0.89 [0.83, 0.94] (p-value = 0.18) f. AUROC comparing predictive performance of Stabl and Lasso on training (left panel) and validation (right panel) cohorts for the COVID-19 dataset. Training: AUROC Stabl = 0.85 [0.74, 0.94] , AUROC Lasso = 0.86 [0.75, 0.94] (p-value = 0.37) . Validation: AUROC Stabl = 0.75 [0.71, 0.79] , AUROC Lasso = 0.76 [0.71, 0.81] (p-value = 0.44) . AUPRC are shown in Fig. S12. g-h. Left panels. Sparsity performances for the PE ( g , number of features selected across cross-validation iterations, median Stabl = 11.0, IQR = [7.8,16.0], median Lasso = 225.5, IQR = [147.5,337.5], p-value < 1e-16 ) and COVID-19 ( h, median Stabl = 7.0, IQR = [4.8,13.0], median Lasso = 19.0, IQR = [8.0,100.0], p-value = 4e-10) datasets. Right panels. Stability path graphs showing the regularization parameter against the selection frequency. The reliability threshold ( θ ), is indicated (dotted line) i-k . Volcano plots depicting the reliability performances of Stabl and Lasso for the PE ( i ), COVID-19 training ( j ) and COVID-19 validation ( k ) datasets. The maximum frequency of selection of each feature is plotted against the −log10 p-value using a univariate Mann-Whitney test. Features selected by Stabl/Lasso only are colored in red/black respectively. Features selected by Stabl are labeled. PE: mean − log10(p-value) Stabl = 8.2; mean − log10(p-value) Lasso = 3.3 . COVID-19 training: mean − log10(p-value) Stabl = 5.5; mean − log10(p - value) Lasso = 5.2 . COVID-19 validation: mean − log(p-value) Stabl = 9.7; mean − log10(p-value) Lasso = 7.8 . Benchmarking of Stabl against elastic net (EN) is shown in Fig. S11 .

Article Snippet: Because clinical omic datasets can vary greatly with respect to dimensionality, signal-to-noise ratio, and technology-specific data preprocessing, we tested Stabl on datasets representing a range of bulk and single-cell omics technologies, including RNA sequencing (RNA-Seq), high-content proteomics (SomaLogic and Olink platforms), untargeted metabolomics, and single-cell mass cytometry.

Techniques: Sequencing, MANN-WHITNEY, Clinical Proteomics, Biomarker Discovery, Selection, Labeling