Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: Power analysis for RNA-Seq differential expression studies using generalized linear mixed effects models

Fig. 1

QQ plots of null LRT and Wald statistics. Data were simulated for 20,000 times at μ=10 and ϕ=1 under the null hypothesis. Sample sizes were set at n=5,10,15,20,and 25. Both the LRT and Wald tests were used for testing mean differences between two conditions under the BNB, Poisson-LMM, or NB-LMM model. Null test statistics are plotted against the Chi-square distribution with 1 degree of freedom. Both the LRT under the BNB model and the LRT under the Poisson-LMM model are approximately following the Chi-square distribution. The Wald test under BNB is slightly above the Chi-square distribution. The LRT under the NB-LMM model is moderately below the Chi-square distribution. Both the Wald test under the Poisson-LMM model and the Wald test under the NB-LMM model are dramatically above the Chi-square distribution

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us