Number Of Samples Per Lane For Multiplexed Rna-Seq

Question

[Dave Bridges]

We are trying to decide how many multiplexed samples to use for a RNA-seq experiment. If we expect ~50M usable reads per lane and presume a 50K transcriptome to get ~1000 reads per gene, then 5 multiplexed samples would give 200 reads per "average" transcript. Is that level of coverage reasonable for differential expression analysis, or should we aim for higher coverage?

[Gww]

Just remember that genes have differing levels of expression; you will get a large proportion of your reads mapping to the most highly expressed genes.

Answer 1

[Istvan Albert]

I would recommend reading some of the relevant publications, a good start would be:

RNA-Seq: a revolutionary tool for transcriptomics

Answer 2

[Travis]

Reads required will be dependent on your desired application/analysis also.

Some numbers recommended by Illumina for human transcriptome sequencing:

3´-SAGE:            1-5 M reads
WT-GE:              5-10 M reads
Alt splicing:       10-50 M reads
Novel Transcripts:  >50 M reads

[Dave Bridges]

is WT-GE normal gene expression for differential comparason?

[Aaron Statham]

Got a link for those numbers?

[Travis]

Aaron, the numbers came directly from a user-group presentation so no link unfortunately. Dave, yes so far as I am aware.