How to filter genes in time course RNA-seq experiments with two replicates for clustering?

Question

[liux.bio]

Hi,

Biostars. I have a time course RNA-seq experiments(7 time points) with two replicates and I want to cluster genes using Mfuzz package. Following the reference and the posts in Biostar, I filter the genes using SD(standard deviation). But I think it's not enough and I want to care about the biological deviations and make use of the replicates. For each time point, define the high value as High.value and the low value as Low.value. All the High.values form a vector and the same as Low.values. Then compute the Pearson Correlation and filter genes with low values. Is this method reasonable?

Additionally, I hope someone give me some suggestions.