lpantano
commented
6 years ago Hi!
sorry about this, maybe you can use the online docs:
http://lpantano.github.io/isomiRs/reference/isoNetwork.html
mi_rse = SummarizedExperiment(assays=SimpleList(norm=mirna_matrix), colData, metadata=list(sign=mirna_keep))That function will work from the mature miRNA only since the databases are only computed using the reference. It is a limitation, but not much to do right now in the bioc world.
What I have done with the experimental researchers, is to down the list to top isomiRs candidates, mainly focusing on trimming and addition events and make experimental assays where you put the sequence in cells and measure the effect on the transcriptome.
Computationally, if you work with mouse or human you can use the custom code from targetscan to predict targets from isomiRs where the seed regions change.
If you have mRNA data, you can make correlations with the isomiRs to detect which ones have a negative relationship and possible detect targets. At some point needs to be some computation to predict the possibility of the isomiRs targeting each gene, but there is not current method for that. That is why I say that people normally move to experimental assays because anyway you need to move there to proof the functionality.
Sorry for the long answer, but there are not a lot of standardization on this.
Cheers
Hi Lorena, I wanted to know about this function 'isoNetwork'.
There are 'mirna_rse' and 'gene_rse' needed for this function but I could not get enough idea of these by the details given in doc. Also , in document it is mentioned how to get these two but it's somehow cut in the pdf at the end and not fully printed. See the image (highlighted) -