defastq

Ultra-fast Multi-threaded FASTQ Demultiplexing

get defastq

download binary

This binary is only for Linux systems: http://opengene.org/defastq/defastq

# this binary was compiled on CentOS, and tested on CentOS/Ubuntu
wget http://opengene.org/defastq/defastq
chmod a+x ./defastq

or compile from source (depend on libdeflate and libisal)

Step 1: download and build libisal

See https://github.com/intel/isa-l_crypto

git clone https://github.com/intel/isa-l_crypto.git
cd isa-l
./autogen.sh
./configure
make
sudo make install

step 2: download and build libdeflate

See https://github.com/ebiggers/libdeflate

git clone https://github.com/ebiggers/libdeflate.git
cd libdeflate
make
sudo make install

step 3: download and build defastq

# get source (you can also use browser to download from master or releases)
git clone https://github.com/OpenGene/defastq.git

# build
cd defastq
make

# Install
sudo make install

If you want to create static build using make static, please make sure that the fold isa-l, libdeflate and defastq are in the same directory.

all options

usage: ./defastq --in1=string --barcode_place=string --index=string [options] ... 
options:
  -1, --in1                   input file name for read1 (string)
  -2, --in2                   input file name for read2 (string [=])
  -b, --barcode_place         For MGI it should be read1 or read2, for Illumina, it should be index1/index2/both_index (string)
  -s, --barcode_start         If barcode_place is read1 or read2, the barcode starting position should be specified. This is 1-based. (int [=0])
  -l, --barcode_length        If barcode_place is read1 or read2, the barcode length should be specified (int [=0])
  -i, --index                 a CSV/TSV/FASTA file contains two values (filename, barcode) (string)
  -r, --reverse_complement    specify this if the index barcodes are reverse complement.
  -o, --out_folder            output folder, default is current working directory (string [=.])
  -u, --undecoded             the file name to store undetermined reads, default is 'undecoded'. To discard the undetermined reads, specify discard (string [=undecoded])
  -z, --compression           compression level for gzip output (1 ~ 12). 0 means no compression, 1 is fastest, 12 is smallest, default is 6.  (int [=6])
  -a, --allowed_mismatch      allowed mismatch (0~2) (int [=0])
  -n, --thread                number of threads (at least 4 for SE, 5 for PE), default 0 means one thread per core. (int [=0])
  -m, --memory                memory limit (GB), 4GB is minimal, default 0 means unlimited. (int [=0])
      --debug                 print debug information.
  -?, --help                  print this message