Create count matrix with different summarizing options

Share:

Description

This function collapses isomiRs into different groups. It is a similar concept than how to work with gene isoforms. With this function, different changes can be put together into a single miRNA variant. For instance all sequences with variants at 3' end can be considered as different elements in the table or analysis having the following naming hsa-miR-124a-5p.iso.t3:AAA.

Usage

1
2
isoCounts(ids, ref = FALSE, iso5 = FALSE, iso3 = FALSE, add = FALSE,
  subs = FALSE, seed = FALSE, minc = 1, mins = 1)

Arguments

ids

object of class IsomirDataSeq

ref

differentiate reference miRNA from rest

iso5

differentiate trimming at 5 miRNA from rest

iso3

differentiate trimming at 3 miRNA from rest

add

differentiate additions miRNA from rest

subs

differentiate nt substitution miRNA from rest

seed

differentiate changes in 2-7 nts from rest

minc

int minimum number of isomiR sequences to be included.

mins

int minimum number of samples with number of sequences bigger than minc counts.

Details

You can merge all isomiRs into miRNAs by calling the function only with the first parameter isoCounts(ids). You can get a table with isomiRs altogether and the reference miRBase sequences by calling the function with ref=TRUE. You can get a table with 5' trimming isomiRS, miRBase reference and the rest by calling with isoCounts(ids, ref=TRUE, iso5=TRUE). If you set up all parameters to TRUE, you will get a table for each different sequence mapping to a miRNA (i.e. all isomiRs).

Examples for the naming used for the isomiRs are at http://seqcluster.readthedocs.org/mirna_annotation.html#mirna-annotation.

Value

IsomirDataSeq object with new count table. The count matrix can be access with counts(ids).

Examples

1
2
3
4
5
6
data(mirData)
ids <- isoCounts(mirData, ref=TRUE)
head(counts(ids))
# taking into account isomiRs and reference sequence.
ids <- isoCounts(mirData, ref=TRUE, minc=10, mins=6)
head(counts(ids))