PEAR - Paired-End reAd mergeR

PEAR is an ultrafast, memory-efficient and highly accurate pair-end read merger. It is fully parallelized and can run with as low as just a few kilobytes of memory.

PEAR evaluates all possible paired-end read overlaps and without requiring the target fragment size as input. In addition, it implements a statistical test for minimizing false-positive results. Together with a highly optimized implementation, it can merge millions of paired end reads within a couple of minutes on a standard desktop computer.

Latest News

PEAR 0.9.8 released


An updated version of PEAR is released.

  • Added --version options
  • Fixed minor error when loading bzip files
  • Lowered autotools version

PEAR 0.9.6 released


An updated version of PEAR is released.

  • Added zlib support
  • Renamed options from --left-fastq and --right-fastq to --forward-fastq and --reverse-fastq, respectively.
  • Added --nbase option that places an N as the result of merging when the base-pair to be merged consists of two different, non-degenerate bases
  • Updated man-pages with the new options

PEAR 0.9.5 released


An updated version of PEAR is released.

  • Replaced assertion when the paired-end read has no overlap but still passes the statistical test

PEAR 0.9.4 released


An updated version of PEAR is released.

  • Support for bzip2 compressed files
  • Simplified code
  • Correct representation of degenerate characters on output (reverseded+complimented)

PEAR 0.9.3 released


An updated version of PEAR is released.

  • Corrections in documentation
  • Added forgotten -c option in getopt_long (fix for capping quality scores in output)

PEAR 0.9.2 released


An updated version of PEAR is released.

  • Added checks whether output files are writable
  • Fixed condition to allow 1 base long reads

PEAR 0.9.1 released


An updated version of PEAR is released.

  • Capped the quality scores of the overlapping region
  • Minor bug fixes

PEAR 0.9 released


An updated version of PEAR is released.

  • PEAR can now process paired-end reads where the lengths of the two ends differ
  • Added some statistics at output

PEAR 0.8.1 released


An updated version of PEAR is released.

  • Fixed numerical issue when computing empirical base frequencies
  • Added a warning when input files differ in number of lines

PEAR Manual released


A basic version of the PEAR manual was added to the website.

Version 0.8 released


The first version of PEAR is finally available. It is available as a standalone library and in source code format.

  • Added memory options
  • Improved explanations in command-line arguments
  • Fixed usage of PHRED base

Important notice

The PEAR creative commons license prohibits commercial use of the code. For testing and using PEAR on a commercial basis you need to purchase a commercial software license. If you wish to purchase such a license please contact:
Prof. Alexandros Stamatakis

PEAR wrapper scripts

Morgan Langille at the Department of Pharmacology at Dalhousie University developed a wrapper perl script for PEAR that facilitates running it on many samples. It also captures the log output and reformats the output to a tab-delimited format (both logs are captured). The script is called and it is available via Morgan's github repository.

To download PEAR please go to the new download page