Hi Uelinton,

I think this just boils down to a question of sequencing depth, since with both technologies, similar read lengths and qualities are expected as far as I know.

I think if you take the effort to prepare 9 libraries each with their own adapters and indices, Hi-Seq is probably the best choice, since you can fit the entire experiment into one run (or two, depending on the desired sequencing depth), eliminating potential sources of between-run error. With myseq, you would have to run the machine multiple times, either with small parts of each sample, which is cumbersome, or with 1 sample = 1 run, which may introduce bias. 

Thanks a lot Rik! I am new to the "sequencing" world and as I noticed, there many options. Concerning the sequencing depth, what would you suggest as being good enough for transcption analysis, considering the genome of appoximately 5 Mb? Adding on that,  there is also the possibility of doing paired ended and just one way sequencing. I guess paired ended sequencing adds on the depth, right?

Hi Uelinton,

Generally the deeper the better, though I don't know whether for simpler transcriptomes like bacteria the added value of deeper sequencing is comparable to eukaryotes.

If you have 9 samples, I guess you could go with one lane (which should give you ~15M sequences per sample if I'm counting correctly).

Paired end is mainly interesting for transcriptome assembly, or the detection of differential splicing or genomic rearrangements. For bacterial gene expression, the added value seems less obvious to me. It does not add to sequencing depth, since you count the number of mappings for your pairs, not for each end of the pair (since they originate from the same fragment, you cannot count them twice).

Great Rik!! Thanks a lot for the clarification!

Cheers