Skip to main content

Table 2 Settings of Scenario 2 to compare the performances regarding joint score estimation, joint loading estimation, and feature selection

From: Statistical integration of two omics datasets using GO2PLS

Methods

GO2PLS; SO2PLS; O2PLS; gPLS; sPLS; PLS

Measure

\(R^2_{\widehat{T}T}\), \(R^2_{\widehat{T}\widehat{U}}\), TPR, \(W^{\top } \widehat{W}\)

p; q

20000; 10000

relevant p; q

1000; 500

N

[30, 100, 200, 300, 600]

\(\sigma ^2_{t_{\perp }} / \sigma _t^2\)

[1/5, 1/3, 1/2, 1, 2, 3, 5]

noise level \(\alpha\)

[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]

  1. \(R^2_{\widehat{T}T}\) and \(R^2_{\widehat{T}\widehat{U}}\) quantify the joint score estimation performance; TPR measures the feature selection performance; \(W^{\top } \widehat{W}\) quantifies the joint loading estimation performance. The dimensions and number of relevant features are set based on the CVON-DOSIS study. Sample size N, the relative strength of orthogonal signal (\(\sigma ^2_{T_{\perp }} / \sigma _T^2\)), and noise level \(\alpha\) are varied
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us