The DeconSeq tool can be used to automatically detect and efficiently remove sequence contamination from genomic and metagenomic datasets. It is easily configurable and provides a user-friendly interface.

References:

• R. Schmieder, R. Edwards
  Fast identification and removal of sequence contamination from genomic and metagenomic datasets.
  PLoS One 2011, 6: e17288, doi: https://doi.org/10.1371/journal.pone.0017288.

• DeconSeq SourceForge page

• Module Name: DeconSeq (see the modules page for more information)
• Unusual environment variables set
  • DECONSEQ_HOME  installation directory
  • DECONSEQ_BIN     executable directory
  • DECONSEQ_SRC    source directory
  Interactive job
  Interactive jobs should be used for debugging, graphics, or applications that cannot be run as batch jobs.

  Allocate an interactive session and run the program. Sample session:

  [user@biowulf]$ sinteractive --mem=8g
  [user@cn3316 ~]$ module load DeconSeq 
  

  1) Using a modified version of BWA, create index databases to be used for contaminat screening. In this example, human and mouse genome databases will be created and stored in the folder db_dir:

  [user@cn3316 ~]$ mkdir db_dir
  [user@cn3316 ~]$ ln -s $DECONSEQ_DATA/hg19_genome.fa 
  [user@cn3316 ~]$ bwa64 index -p hg19_db -a bwtsw hg19_genome.fa >out.txt 2>&1 &
  [user@cn3316 ~]$ mv hg19_db.* db_dir
  [user@cn3316 ~]$ ln -s $DECONSEQ_DATA/mm10_genome.fa 
  [user@cn3316 ~]$ bwa64 index -p mm10_db -a bwtsw mm10_genome.fa >out.txt 2>&1 &
  [user@cn3316 ~]$ mv mm10_db.* db_dir
  [user@cn3316 ~]$ ls db_dir
  hg19_db.amb  hg19_db.rbwt  mm10_db.amb  mm10_db.rbwt
  hg19_db.ann  hg19_db.rpac  mm10_db.ann  mm10_db.rpac
  hg19_db.bwt  hg19_db.rsa   mm10_db.bwt  mm10_db.rsa
  hg19_db.pac  hg19_db.sa    mm10_db.pac  mm10_db.sa
  

  2) Download the configuration file DeconSeqConfig.pm from $DECONSEQ_SRC and modify it as needed, or simply download from $DECONSEQ_DATA the file already modified for the human and mouse databases:

  [user@cn3316 ~]$ cp $DECONSEQ_DATA/DeconSeqConfig.pm . 
  

  3) Get/prepare sample data file(s). In this small example, idata from first 1000 reads of the file SRR4254643_mouse.fasta will be used:

  [user@cn3316 ~]$ ls -s $DECONSEQ_DATA/SRR4254643_mouse.fasta 
  [user@cn3316 ~]$ head -n 1000 SRR4254643_mouse.fasta > sample.fasta
  

  4) Finally, run deconseq on the data file and get the results:

  [user@cn3316 ~]$ deconseq -f  test.fasta -dbs hsref  -dbs_retain mouse
  

  It order to view other available deconseq commands, type:

  [user@cn3316 ~]$ deconseq -h
  

  Batch job
  Most jobs should be run as batch jobs.

  In order to process a large data file on cluster, first split it into smaller chanks, e.g.

  [user@cn3316 ~]$ splitFasta -verbose -i SRR4254643_mouse.fasta -s 2     #chunks of 2MB
  [user@cn3316 ~]$ splitFasta -verbose -i SRR4254643_mouse.fasta -n 10    #10 chunks
  

  and then run the deconseq command separately on each the chunk.
  

  To this end, create a batch input file, e.g. deconseq.sh:

  #!/bin/bash
  module load DeconSeq
  ln -s $DECONSEQ_DATA/db_dir
  ln -s $DECONSEQ_DATA/SRR4254643_mouse.fasta
  splitFasta -verbose -i SRR4254643_mouse.fasta -n 10
  deconseq -f  SRR4254643_mouse.fasta_c1.fasta  -dbs hsref  -dbs_retain mouse
  deconseq -f  SRR4254643_mouse.fasta_c2.fasta  -dbs hsref  -dbs_retain mouse
  ...
  deconseq -f  SRR4254643_mouse.fasta_c10.fasta -dbs hsref  -dbs_retain mouse
  

  Submit this job using the Slurm sbatch command.

  sbatch [--cpus-per-task=#] [--mem=#] deconseq.sh

  HPC @ NIH ~ Contact

  Disclaimer ~ Privacy ~ Accessibility ~ CIT ~ NIH ~ DHHS ~ USA.gov