novoalab / EpiNano

Detection of RNA modifications from Oxford Nanopore direct RNA sequencing reads (Liu*, Begik* et al., Nature Comm 2019)
GNU General Public License v2.0
110 stars 31 forks source link

Hello #5

Closed ishaaq34 closed 5 years ago

ishaaq34 commented 5 years ago

How much coverage do we need for a direct RNA sequencing experiment that would be optimum for identification of m6a sites ?

enovoa commented 5 years ago

It will depend on the expression levels of the gene, as well as on the stoichiometry of the modification at a given site. If the site is only modified at low levels and you have poor coverage, you won't be able to detect anything. We have not benchmarked the optimum coverage needed, as in our cases we were using high coverage data from training/testing. I would recommend not going below 50reads coverage per site, but as mentioned, we have not done any type of benchmarking to determine this number.

enovoa commented 5 years ago

However, it is something that could be done as improvement as I agree that your question is relevant to apply the software to unknown scenarios. Thanks for your contribution

ishaaq34 commented 5 years ago

Thanks for the response ,

We are interested in comparing the m6a topology of infected and control transcriptome of a animal through nanopore . What special things you would like to suggest ?

ishaaq34 commented 5 years ago

As we know m6a is most prevalent modifications !!

enovoa commented 5 years ago

Detecting m6A modifications in animals using nanopore will likely require either the use of improved protocols to improve the throughput, the use of many flowcells (expensive), or the use of promethion direct RNA sequencing. So unless you have access to promethion sequencing, I think that it might be a bit early to move forward with such design, but I could be wrong. Best of luck!

ishaaq34 commented 5 years ago

Thanks for your Response,

Lets us talk about only mRNAs not whole transcriptome How many flow cells will required to get modifications of mRNA identified (per sample)

enovoa commented 5 years ago

Hi ishaaq, I believe that this type of question would be best that you would address to Nanopore using the community portal. We cannot provide expertise on how many flowcells you will need for your specific biological questions or organisms. We are currently working on this with our own samples and cannot provide an answer for your specific experiment.