[Jane M]

Ok, thank you.

Thank you everybody for your answers.
Do not hesitate to tell here what you do, what are your choices. It would be nice to see if there is a consensus about those questions, when not limited by the cost.

[gringer]

It would be nice to see if there is a consensus about those questions, when not limited by the cost.

I'll give you my take on depletion vs enrichment, as a mostly theoretical bioinformatician. Taking a high-level view, there are two ways to filter rRNA:

• select/tag the rRNA sequences and remove them [depletion]
• select/tag the non-rRNA sequences and extract them [enrichment]

Depletion is less likely to introduce bias, but also less likely to be successful in removing rRNA -- it's difficult to tag all rRNA seqences due to degredation, mutation, etc.. Enrichment introduces bias, because you need to know in advance how to enrich the sequences that you are interested in (e.g. polyA for mRNA enrichment), but will give you better resolving power (i.e. fewer total reads required for the same power) because you can know for sure that you're excluding rRNA in your enrichment process.