GitHub - ETHZ-INS/scanMiR: A set of tools for scanning for miRNA binding sites and working with affinity models (original) (raw)

scanMiR

scanMiR is a bioconductor set of tools for working with miRNA affinity models (KdModels), enabling efficient and flexible scanning for miRNA binding sites and prediction of target repression.

This repository details the R package; for the web interface click here.

Getting started

Installation

BiocManager::install("ETHZ-INS/scanMiR")

Basic example usage

# accepts miRNA target seeds:
findSeedMatches(seqs, seeds="AAACCAC")
# full miRNA sequences:
findSeedMatches(seqs, seeds="UUAAUGCUAAUCGUGAUAGGGGUU")
# or KdModels:
findSeedMatches(seqs, seeds=model)

GRanges object with 43 ranges and 4 metadata columns:
       seqnames    ranges strand |  p3.score          type    log_kd     note
          <Rle> <IRanges>  <Rle> | <integer>      <factor> <integer>    <Rle>
   [1]     seq2 2687-2694      * |         4 7mer-a1           -3607        -
   [2]     seq2 2358-2365      * |         0 6mer              -2341        -
   [3]     seq2 2550-2557      * |         0 6mer-m8            -986        -
   [4]     seq2 1642-1649      * |         0 non-canonical      -952        -
   [5]     seq2   920-927      * |         0 non-canonical      -847        -
   ...      ...       ...    ... .       ...           ...       ...      ...

The putative 3' binding of each match can also be visualized:

viewTargetAlignment(matches[1], miRNA=model, seqs=seqs)

miRNA            3'-UUGGGGAUAGUGCUA---AUCGUAAUU-5'     
                              |||||     ||||||-        
target 5'-...NUAUAGACGAGUGACCUACGAUAUGCCGCAUUAAUU...-3'

scanMiR can predict TDMD and circRNA slicing sites, and aggregate sites to predict repression based on the biochemical model fromMcGeary, Lin et al. (2019). For more information, see our paper.

To learn more about the functionalities, see the package'svignettes. To obtain predicted KdModels for all mouse, human and rat miRbase miRNAs, see thescanMiRData pacakge.

For convenient wrappers connecting to AnnotationHub, fast out-of memory access to large collections, or a web interface to scanMiR, see thescanMiRApp package.