Fluseq

A bioinformatics pipeline for influenza virus genome sequencing that includes contig assembly, blastn, filter for in-run, between-sample contamination reads, and consensus sequence calling.

View project on GitHub

FluSeq-v1.0

Version 1.0 for Illumina Miseq system. For Research Use only. Not for use in diagnostic procedures.

FluSeq is a bioinformatics program developed and tested for an influenza A genome-wide amplicon-based high-throughput sequencing (HTS) method, using Illumina Miseq system [1]. It includes contigs assembly, blastn, filters for in-run contamination reads, and consensus sequence calling. Users are advised to refer the end-to-end laboratory protocol described by Lee, et al. 2016 [1]. Users are encouraged to modify the FluSeq program written in python scripts for other clinical virus amplicon-based HTS methods.

INSTALLATION

Installation instructions are available for LINUX/UNIX or MacOSX only. The entire analytic workflow was implemented and tested by the author on CentOS-6.6/RedHat Linux and MacOSX, but not on other operating systems. The operating system should contain the latest version of java.

Install Python 3.4.3 and pip

(https://www.python.org/downloads/)

  • Upon installation, log in as root from Terminal (Linux and MacOSX): 
# pip3.4 install pandas
# pip3.4 install numpy
# pip3.4 install ZODB
# pip3.4 install lxml 
# pip3.4 install xlrd
# pip3.4 install xlwt
# pip3.4 install beautifulsoup4
# pip3.4 install scipy (# yum install lapack-devel if lapack/blas resources not found in the system)
# pip3.4 install transaction

Install VICUNA-v1.3

(http://www.broadinstitute.org/scientific-community/science/projects/viral-genomics/vicuna)

  • The vicuna_config.txt used by this analytic workflow is stored in FluSeq Folder. Compilation of vicunAnalysis is not required during the installation.
  • Path to the compiled vicuna executive file needs to be changed accordingly in the FluSeq-v1.0.py, i.e. Line 100: First argument of the subprocess.Popen, to “YourPathInstalled/VICUNA_v1.3/bin/vicuna-omp-v1.0”
  • Optional: Line 104 - logging.info: Change program path to “YourPathInstalled/VICUNA_v1.3/bin/vicuna-omp-v1.0”.

Install BLAST 2.2.31+ software

(https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastDocs&DOC_TYPE=Download)

Install BWA-0.6.2 software

(http://sourceforge.net/projects/bio-bwa/files/)

  • Users are advised to install BWA-0.6.2 software but not any version after this, as bwa-aln/sampe in this version was found to be more stable in reads alignment for this study.

Install samtools-1.2

(http://sourceforge.net/projects/samtools/files/samtools/1.2/)

Install GATK-3.4-46 software

(https://www.broadinstitute.org/gatk/download/)

  • Path to the GenomeAnalysisTK.jar executive file needs to be changed accordingly in the FluSeq-v1.0.py, i.e. Lines 233, 259, and 269: Third or Fourth argument of the subprocess.Popen, to “YourPathInstalled/GenomeAnalysisTK.jar”.
  • Optional: Lines 243, 265, and 278: logging.info: Change program path to “YourPathInstalled/GenomeAnalysisTK.jar”.

Install picard-tools-1.138

(https://github.com/broadinstitute/picard/releases/tag/1.138)

  • Path to the GenomeAnalysisTK.jar executive file needs to be changed accordingly in the FluSeq-v1.0.py, i.e. Line 146: Third argument of the subprocess.Popen, to “YourPathInstalled/picard-tools-1.138/picard.jar”.
  • Optional: Line 152: logging.info: Change program path to “YourPathInstalled/picard-tools-1.138/picard.jar”.

Install FluSeq

a. Download the FluSeq.tar, decompress, and mv the FluSeq Folder to ~.

$ tar –xvf FluSeq.tar
$ mv FluSeq ~

b. Change working directory to FluSeq 

$ cd ~/FluSeq

USAGE

Influenza Sequence Database

This section describes the sequence database used for blastn. It contains all influenza sequences available in GenBank, updated to the date accessed by the user.

Creating/Updating Sequence Database

  1. Download all influenza sequences available from NCBI Influenza Virus Resource with a customized FASTA defline as “>{accession}|{strain}|{segment}|{serotype}” and save the fasta file as FASTA.fa in FASTADatabase directory.
  2. Change working directory to FASTADatabase
$ cd ~/FluSeq/FASTADatabase
  1. Execute DBupdate
$ ./DBupdate.py FASTA.fa
or
$ python3 DBupdate.py FASTA.fa
or
$ python3.4 DBupdate.py FASTA.fa

FluSeq INPUT FILES and FOLDER

This section describes the input FluSeq-v1.0 requires.

Folder and File Naming

The top-level input folder can be named in any combination of alphanumeric characters. It is advisable that the first letter of the name of a folder is in upper case for a better folder and file organization, e.g. Folder20151231 (Figure 1 – Folder in red).

The input folder should contain paired-end reads in fastq.gz format generated by Miseq Reporter or fastq2bcl2, e.g. SampleID_L001_R1_001.fastq.gz and SampleID_L001_R2_001.fastq.gz (Figure 1). Also, it should contain the ResequencingRunStatistics.xml that can be copied from the top-level run folder named according to Miseq , e.g. YYMMDD_machinename_experimentnumber_flowcellnumber.

Figure 1

Figure 1. Directory of FluSeq and input folder. Input folders for run analysis and contamination learning are colored in red and orange, respectively.

Contamination Learning (Optional)

The current FluSeq software provides the database file for MS2 contamination statistics (i.e. ms2SysCtrl.db). However, it is recommeded that individual laboratories generate a laboratory-specific database file to allow more accurate statistics.

The input folder for in-run contamination learning should contain paired-end reads for an in-run negative system control (NSC) in fastq.gz format generated by Miseq Reporter or fastq2bcl2, e.g. SampleID_L001_R1_001.fastq.gz and SampleID_L001_R2_001.fastq.gz (Figure 1 – Folder in orange). Also, it should contain the ResequencingRunStatistics.xml that can be copied from the top-level run folder named according to Miseq , e.g. YYMMDD_machinename_experimentnumber_flowcellnumber.

The database will be updated according for each execution. A new database file will be generated if the ms2SysCtrl.db file is not found in the FluSeq folder.

Procedure

  1. Change working directory to FluSeq 
$ cd ~/FluSeq
  1. Execute FluSeq
$ ./ms2ContLearning.py InputFolder 151206
or
$ python3 ms2ContLearning.py InputFolder 151206
or
$ python3.4 ms2ContLearning.py InputFolder 151206

Execution of FluSeq analysis

a. Change working directory to FluSeq 

$ cd ~/FluSeq

b. Execute FluSeq

$ ./FluSeq-v1.0.py InputFolder
or
$ python3 FluSeq-v1.0.py InputFolder
or
$ python3.4 FluSeq-v1.0.py InputFolder

REFERENCE:

[1] Hong Kai Lee, Chun Kiat Lee, Julian Wei-Tze Tang, Tze Ping Loh, and Evelyn Siew-Chuan Koay. Contamination-controlled high-throughput whole genome sequencing for influenza A viruses using the MiSeq sequencer. Sci Rep. 2016, 6:33318. doi: 10.1038/srep33318.