A Practical Approach to Calculating Sample Size Based on Generalized Linear Models for Assessing Differential Expression Analysis in RNA-Seq Data

• Activity Number: 145
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Chung-I Li*
• Companies: National Chiayi University
• Keywords: Sample size ; RNA-seq ; false discovery rate ; generalized linear model ; exemplary dataset

Abstract

As RNA-seq rapidly develops and costs continue to decrease, more samples will continue to be sequenced. Thus, the determination of sample size becomes an important issue in planning the experimental design. Some current methods for calculating the required sample size of a study are based on the hypothesis testing framework, assuming the counts for each gene come from the Poisson or negative binomial distributions. However, these methods are limited in terms of accommodating covariates. To deal with this issue, we propose an estimating procedure based on the generalized linear model. By constructing a representative exemplary dataset and estimating the conditional power, the proposed method is easy to use and requires no complicated mathematical approximations or formulas. Most attractively, the downstream analysis can be based on the many currently existing R/Bioconductor packages. Finally, the proposed method is applied to two real-world studies.