Before starting a sequencing run, you need to know the depth of sequencing you want to achieve. Researchers typically determine the amount of coverage needed based on experimental requirements, genes being examined and their expression levels, the reference genome and published literature.

We’ve recently published a guide to serve as a starting point for those trying to determine how deep they should sequence their samples:

Coverage and Read Depth Recommendations by Application

Much of the data in this table comes from published coverage saturation experiments where depth was compared to another specific metric, e.g. differential expression.

It should be noted that increasing sequencing depth is not always the best solution.  Several studies have demonstrated that more replicates in RNA-seq  is preferred over increasing the number of sequence reads (1, 2). This is nicely illustrated by Liu, Y et al., 2014 in the figure below. An increase in biological replicates, from 2 to 7 significantly increases the number of identified differential expressed genes whereas increasing sequencing reads past 10 million has diminishing returns. 

Replicates Versus Sequencing Depth 

Once you’ve determined the coverage you need, calculate the number of sequencing reads required to achieve that coverage using Genohub’s coverage and read calculator. Send us a consultation request if you need help trying to determine the coverage required in your experiment. 

 Citations: 

1-     Liu Y, Zhou J, White KP. RNA-seq differential expression studies: more sequence or more replication? Bioinformatics. 2014 Feb 1;30(3):301-4

2-     Rapaport F, Khanin R, Liang Y, Krek A, Zumbo P, Mason CE. Comprehensive evaluation of differential expression analysis methods for RNA-seq data. Genome Biology 2013, 14:R95 2013