sRNAbench is a free web-server tool and standalone application for processing small-RNA data obtained from next generation sequencing platforms such as Illumina or SOLiD. The sRNAbench tool is the replacement for miRanalyzer.

sRNAbench is now included as part of sRNAtoolbox

Main features

  • Two different ways can be used to profile the expression levels of small RNAs depending on whether a good genome sequence/annotation is available: i) mapping of all short sequence reads against the genome, obtaining the expression levels by means of annotations in fasta or bed format and ii) mapping against sequence libraries in fasta/Bowtie index format (like done by miRanalyzer).
  • An unlimited number of genomes can be used in the analysis at the same time without having to pool all sequences into a single file/Bowtie index. This feature is specially important when analyzing the interaction between parasites and hosts, symbiosis or virus infected cells.
  • Adapter trimming can be performed and sRNAbench accepts fastq, fastq.gz, sra, read count and fasta input format
  • Extensive profiling of all microRNA sequence and length variants (isomiRs). Furthermore, NTAs (non-templated additions) can be detected for all sequence libraries and not only for microRNAs.
  • Several summary and graphical summaries are available.
  • The prediction of novel microRNAs was improved compared to miRanalyzer in the sense that it is much more specific now. The prediction is based on structural, sequence and biogenesis features.
  • Detection of differentially expressed microRNAs (including newly predicted ones)



20/04/2019: Web-server update to miRBase version 22

27/08/2014: Web-server update to miRBase version 21 & posibility to analyse high-confidence set (see

Release notes:

The different versions will be distinguished by  ‘Month/Year’. For example, the first released version is 12/13. You can launch the program with ‘-v’ on the command line to get the version.

07/10/2014: Version 10/14 released: current sRNAbench.jar

21/07/2014: Version 07/14 released: 07/14 sRNAbench.jar

12/05/2014: Version 05/14 released:

16/12/2013:  Version 12/13 released: 12/13 sRNAbench.jar

Bug fixes /changes


  • Barcode trimming didn’t work for ‘gz’ files
  • Prediction of novel microRNAs changed: two sets are defined now; high confidence and low confidence (see manual for details)
  • Homologous name assignment improved
  • Error occurred if an adapter trimmed fastq input with ‘empty reads’ (length 0) was provided: fixed


  • Rewrite some functions to make the program faster
  • Prediction of novel microRNAs: fixed bug in the selection of the correct novel microRNAs out of several overlapping candidates
  • Read length distribution for genome mapped reads


  • Length distribution of mapped reads (genome mapped & library mapped)
  • Global length distribution: untrimmed reads had been omitted (fixed)
  • Prediction of novel microRNAs: fixed wrong read count for candidates at the (-) strand

Download and install the database

sRNAbench relies on a local database where the library files, genome sequences and Bowtie indexes need to be stored. The database can have any arbitrary name and the easiest way to generate it is by means of the “start-up” DB following the next steps:

  1. Download the “start-up” database:
  2. Extract it to the directory of your choice: tar xvzf sRNAbenchDB.tgz (for example “/home/user”)
  3. Download the most recent version and replace the sRNAbench.jar file form the database: current sRNAbench.jar
  4. Download the most recent version of the makeSeqObj.jar: current makeSeqObj.jar
  5. Only for DE: Download the most recent version of the differential expression module sRNAbenchDE.jar and place it into the database: current sRNAbenchDE.jar
  6. Only for DE: R, bioconductor and edgeR need to be installed
  7. Only for DE: for some analysis types you will need the The Apache Commons Mathematics Library


Please cite

If you use the sRNAbench webserver for your publication please cite the following publications:


Aparicio-Puerta, Ernesto, et al. (2019) sRNAbench and sRNAtoolbox 2019: intuitive fast small RNA profiling and differential expression. Nucleic acids research

Barturen et al. (2014) sRNAbench: profiling of small RNAs and its sequence variants in single or multi-species high-throughput experiments. Methods in Next Generation Sequencing.


Ben Langmead, Cole Trapnell, Mihai Pop and Steven L Salzberg (2009) Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biology


miRBase: integrating microRNA annotation and deep-sequencing data.
Kozomara A, Griffiths-Jones S. NAR 2011 39 (Database Issue):D152-D157

Furthermore, if you use data derived from our Helper Tools please cite the publications given at the data retrieval page.


Frequently asked questions


Q: The sum of 5p and 3p arms from the miRBase_main.txt file does not give the same number as the sum from the mature_sense.grouped file

A: The miRBase_main.txt file contains only those microRNAs, for which no inconsistencies have been detected. Inconsistencies can be: no stem-loop, at least one mature sequence folds back onto itself in the secondary structure.


Q: The number of detected pre-microRNAs given by the program does not fit the sum of entries in the hairpin_sense.grouped and hairpin_antisense.grouped files.

A: The number of detected pre-microRNAs given by sRNAbench is the number of pre-microRNAs (hairpin sequences) mapped in sense or antisense orientation. hairpin_sense.grouped and hairpin_antisense.grouped files can contain ‘duplicates’ as a microRNA gene might have more than one copy in the genome. Furthermore, the hairpin_antisense.grouped file will contain also pre-microRNAs already listed in the hairpin_sense.grouped file. Therefore, the number of detected pre-microRNAs given by sRNAbench is the number of pre-microRNAs detected in sense PLUS the number of pre-microRNAs mapped uniquely in antisense direction.