Skip to content


Folders and files

Last commit message
Last commit date

Latest commit


Repository files navigation

Flexbar – flexible barcode and adapter removal

The program Flexbar preprocesses high-throughput sequencing data efficiently. It demultiplexes barcoded runs and removes adapter sequences. Several adapter removal presets for Illumina libraries are included. Flexbar computes exact overlap alignments using SIMD and multicore parallelism. Moreover, trimming and filtering features are provided, e.g. trimming of homopolymers at read ends. Flexbar increases read mapping rates and improves genome as well as transcriptome assemblies. Unique molecular identifiers can be extracted in a flexible way. The software supports data in fasta and fastq format from multiple sequencing platforms.

Refer to the manual or contact Johannes Roehr for support with this application.


Johannes T. Roehr, Christoph Dieterich, Knut Reinert:
Flexbar 3.0 – SIMD and multicore parallelization. Bioinformatics 2017.

See article on PubMed

Matthias Dodt, Johannes T. Roehr, Rina Ahmed, Christoph Dieterich:
Flexbar – flexible barcode and adapter processing for next-generation sequencing platforms. Biology 2012.

See article on PubMed

Flexbar logo


Flexbar source code as well as binaries for Linux and Mac OS can be downloaded on the release page. Please follow instructions for building or setup of binaries below. Additionally, Flexbar is available via package manager on Debian systems, in Homebrew, and in Bioconda. Versions before 2.4 can be found on the old page.

Installation with package managers:

  • Debian: sudo apt install flexbar
  • Homebrew: brew install brewsci/science/flexbar
  • Bioconda: conda install -c bioconda flexbar

To get the latest version and best performance consider to build Flexbar from source.

Building from source

Make sure that cmake is available, as well as development and runtime files of the TBB library 4.0 or later (Intel Threading Building Blocks). For example on Debian systems, install the packages libtbb-dev and libtbb2. Furthermore, the SeqAn library and a compiler that supports C++14 is required:

  • Get SeqAn library version 2.4.0 here
  • Download Flexbar 3.5.0 source code release

Decompress both files:

tar xzf flexbar-3.5.0.tar.gz
tar xJf seqan-library-2.4.0.tar.xz

Move SeqAn include folder to Flexbar:

mv seqan-library-2.4.0/include flexbar-3.5.0

Use these commands for building:

cd flexbar-3.5.0
cmake .

Flexbar versions from 3.0 up to 3.2 require SeqAn 2.2.0 instead. Flexbar version 2.7 uses SeqAn 2.1.1 and releases prior to 2.7 use the SeqAn 1.4.2 library.


For execution of provided Flexbar binaries, the corresponding TBB library has to be available. Downloads contain the library file for runtime. Follow the platform specific instructions below.


Adjust lib search path to include the absolute path of the Flexbar directory containing the lib file for the current terminal session, or permanently in shell startup scripts:

export LD_LIBRARY_PATH=/YourPath/flexbar-3.5.0-linux:$LD_LIBRARY_PATH

Mac OS

It applies the same as for Linux. Make the file libtbb.dylib available by setting the lib search path:

export DYLD_LIBRARY_PATH=/YourPath/flexbar-3.5.0-macos:$DYLD_LIBRARY_PATH

Program usage

Flexbar needs at least one file with sequencing reads in fasta or fastq format as input. Additionally, the target name and further options can be specified. For read separation based on barcodes and for adapter removal, a file in fasta format with barcode or adapter sequences should be provided.

flexbar -r reads [-b barcodes] [-a adapters] [options]

Refer to the help screen flexbar -h or manual for more information. Although default parameters of Flexbar are optimized to deliver good results in many scenarios, the adjustment of parameters like --adapter-min-overlap might improve results. For tests, run within the test folder if flexbar is reachable via the path variable.

Quality-based trimming

In this example, reads in fastq format are trimmed based on their quality scores in Illumina version 1.8 format. The TAIL method trims the right end of reads until a quality score equal or higher than the threshold is reached, default 20. Trimmed reads are written to target.fastq in same format as the input.

flexbar -r reads.fq -t target -q TAIL -qf i1.8

Demultiplexing with barcodes

Reads that are barcoded on the left end are demultiplexed by specifying a file with barcodes in fasta format. Reads that can be assigned are written to separate files using file names that are based on the names of barcodes in the fasta file.

flexbar -r reads.fq -b barcodes.fa -bt LTAIL

Adapter removal single-end

To remove adapter sequences from single-end reads, specify a file with adapters in fasta format. These are removed from the right side of reads per default, if they do not align before the read start. The left side of reads is kept if long enough. The overlap of an adapter and read must have at least length 3 with at most 10% errors in default settings.

flexbar -r reads.fq -a adapters.fa -ao 3 -ae 0.1

Adapter removal paired-end

For paired-end libraries, specify both files with paired reads and a fasta file with adapters for removal. Given adapters are trimmed in right mode per default. It is recommended to activate the pair overlap detection in case of standard paired reads. This increases the sensitivity by removing very short parts of adapters if an overlap is detected for a pair.

flexbar -r r1.fq -p r2.fq -a a1.fa -a2 a2.fa -ap ON

Adapter removal presets

Several adapter presets for Illumina libraries are included in Flexbar. For example, select the TruSeq preset for standard TruSeq adapters and specify two read files for paired reads. If a preset is chosen, a separate file with adapters is not needed for removal. It is recommended to turn on the pair overlap detection for standard paired-end libraries.

flexbar -r r1.fq -p r2.fq -aa TruSeq -ap ON

For further examples visit the manual page.