Assessing the size of gene or RNAi effects in multifactor high-throughput experiments

Aims: To expand the recently proposed contrast variable and associated concepts to assess the size of gene effects or siRNA effects in multifactor high-throughput experiments, as well as to serve the need to consider both mean and standardized mean of contrast (SMC). Methods & results: The recently proposed concepts of contrast variable and SMC are expanded in the context of multifactor analysis of variance. Based on this expansion, SMC is explored as a tool for analyzing multifactor high-throughput data, a novel plot termed a dual-flashlight plot is proposed, and the incompatibility of false-discovery rates across experiments is demonstrated. The applications show that the results reached using expanded SMC and the dual-flashlight plot are more reasonable than those reached using p-value-based or false-discovery rate-based volcano plot for assessing differential expression, genetic dominance and linear/quadratic time-course changes. Conclusion: Compared with traditional contrast analysis, the expanded contrast variable and SMC may serve as an alternative that can address the real need of assessing the size of gene or siRNA effects in multifactor high-throughput experiments.