Shown is the type 1 error rate (at significance threshold of 0.05) of the five different TWAS models and LAVA-rG for the null hypothesis of no genetic covariance (cov(GE, GY) = 0), at different levels of local heritability (h2) for outcome phenotype (horizontal axis) and gene expression (separate panels). Results are shown for sample sizes of 1,000 and 100,000 for the expression and outcome respectively, with 1,000 SNPs. As shown, the results are very similar across the different models. The lower type 1 error rate for Elastic Net and LASSO compared to the other TWAS models for the 1% expression heritability condition is due to these two models sometimes failing to converge at very low genetic signal for the expression (see also Fig D in S1...
<p>Each panel compares the average error rate between CLAPPER and competing methods for a particular...
Estimates provided over various confounding simulation scenarios. General simulation parameters: gen...
<p>Average type 1 error rates of the statistic for testing interaction between two genes with margin...
Shown is the type 1 error rate (at significance threshold of 0.05) of the LAVA-TWAS model for the nu...
Shown is the type 1 error rate inflation of the LAVA-TWAS model for the null hypothesis of no geneti...
Contains additional derivations and explanation of the core properties of TWAS and related tests and...
For many multivariate twin models, the numerical Type I error rates are lower than theoretically exp...
See S1 Table 1a and 1b for detailed settings. Summary-level association statistics are simulated for...
When evaluating a new statistical test, it’s important to check the type 1 error (T1E) control, ofte...
When evaluating a new statistical test, it’s important to check the type 1 error (T1E) control, ofte...
<p>“Lasso” represents model (2), “Linear low-rank” represents model (3), and “Nonlinear low-rank” re...
<p>(A) Distribution of SNP-based heritability estimates. The horizontal lines indicate the true heri...
<p>* Multiple Testing Correction: performed via Benjamini-Yekutieli procedure with a 5% threshold.</...
<p>Each * indicates one target gene. X-axis represents the test RMSE of the Lasso model, whereas Y-a...
<p>Average type 1 error rates of the statistic for testing interaction between two genes with margin...
<p>Each panel compares the average error rate between CLAPPER and competing methods for a particular...
Estimates provided over various confounding simulation scenarios. General simulation parameters: gen...
<p>Average type 1 error rates of the statistic for testing interaction between two genes with margin...
Shown is the type 1 error rate (at significance threshold of 0.05) of the LAVA-TWAS model for the nu...
Shown is the type 1 error rate inflation of the LAVA-TWAS model for the null hypothesis of no geneti...
Contains additional derivations and explanation of the core properties of TWAS and related tests and...
For many multivariate twin models, the numerical Type I error rates are lower than theoretically exp...
See S1 Table 1a and 1b for detailed settings. Summary-level association statistics are simulated for...
When evaluating a new statistical test, it’s important to check the type 1 error (T1E) control, ofte...
When evaluating a new statistical test, it’s important to check the type 1 error (T1E) control, ofte...
<p>“Lasso” represents model (2), “Linear low-rank” represents model (3), and “Nonlinear low-rank” re...
<p>(A) Distribution of SNP-based heritability estimates. The horizontal lines indicate the true heri...
<p>* Multiple Testing Correction: performed via Benjamini-Yekutieli procedure with a 5% threshold.</...
<p>Each * indicates one target gene. X-axis represents the test RMSE of the Lasso model, whereas Y-a...
<p>Average type 1 error rates of the statistic for testing interaction between two genes with margin...
<p>Each panel compares the average error rate between CLAPPER and competing methods for a particular...
Estimates provided over various confounding simulation scenarios. General simulation parameters: gen...
<p>Average type 1 error rates of the statistic for testing interaction between two genes with margin...