BPGA Files

Older versions are not available [Do not download].
Please visit Documentation Page : https://sourceforge.net/p/bpgatool/wiki/Home/
Latest Recommended Version - BPGA-1.3.0

UPDATE HISTORY:

12.05.2017 : Introduced box plot option from matrix input, under same version name.

PREREQUISITES

    -Windows Requirements:
    System: Windows XP or latter
    Usearch: Get it from http://www.drive5.com/usearch/. Download and rename the Windows executables to "usearch.exe" [case sensitive, also mind that Windows file extensions are visible]
    gnuplot: Download Gnupot_Win-64bit_Version or Download_Gnupot_Win-32bit_Version
    WARNING (for Windows only) : Please check “vcomp100.dll” system file in the system32 or system64 folder, path of this folder is as “C:\Windows\System32”. If not present copy it into above said folder. This file is required for USEARCH to function properly. This dll is available at this link

    -Linux Requirements:
    Usearch : Get it from http://www.drive5.com/usearch/. Download and rename the Linux executable to "usearch". [case sensitive]
    gnuplot: run sudo apt-get install gnuplot-4.6.6 or simply sudo apt-get install gnuplot
    ghostscript: run sudo apt-get install ghostscript
    wine (Ubuntu): sudo add-apt-repository ppa:ubuntu-wine/ppa -y && sudo apt-get update && sudo apt-get install wine
    WARNING (for Debian only) : Make sure you have 'glibc 2.15' or higher (the C library in Debian).If not, you can carefully install higher version of glibc in parallel to 'glibc 2.14' or less. [Do not try to remove original glibc, as other binaries may not work properly.] As the post on one of the debian forum says: "One can install NEW glibc in parallel to OLD in some different place, e.g., in /opt. In fact, this is a common technique to make Google Chrome work on CentOS. I'm not saying this is as easy as 1-2-3, but it is certainly both doable and, if done properly, safe."
    Alternatively, the steps are discussed here for running new applications on old glibc.
    For BPGA Version 1 (Linux) only: Make sure that libgif4 (library) and libtiff4 are installed. Install 'libtiff4' and 'libgif4' from Ununtu Software Center or install manually as follows. Ubuntu 14.04 LTS 64 bit release includes 'libtiff5' library, BPGA may not start execution with it. (BPGA Version 1 currently needs 'libtiff4' and 'libgif4'). You may also get libtiff4 for different Ubuntu Releases from this link and libgif4 from this link. Install them manually using following commands (type exact file name that you downloaded) sudo dpkg -i ./libgif4_version_details_32_or_64_bit.deb and sudo dpkg -i ./libtiff4_version_details_32_or_64_bit.deb

RUN BPGA

    -On Windows:
    Install BPGA by simply double clicking on Installer file or extracting from zip.
    Open BPGA folder and change directory to bin folder. Copy 'usearch.exe' to bin folder.
    Run BPGA-Version-1.exe from bin folder for pan-genome analysis.

    -On Linux:
    Extract the files from tar.gz by command:tar -zxvf bpga-version-X-linux-xxx-xx.tar.gz
    Open BPGA folder and change directory to bin folder. Copy 'usearch' file to bin folder.
    cd to bin and use commands:chmod +x BPGA-Version-XX then ./BPGA-Version-XX
    Note : xxx-xx stands for respective version. (should match which version you downloaded).

ANALYSIS OPTIONS

    1. INPUT PREPARATION :
    Modifies raw .gbk or Protein FASTA files in a format given below, prior to pan genome analysis.
    2. DEFAULT PAN-GENOME ANALYSIS :
    Runs default pan-genome analysis on modified input files, followed by advanced analysis options.
    3. ONE CLICK MODE :
    This option includes the both 1 and 2 options together for whole pan-genome analysis using dafault settings (user friendly) but for more than 100 genomes this will not include subset analysis and KEGG/COG functional analysis.

    -Important:
	To use BPGA with cd-hit and OrthoMCL outputs, follow these steps:
    1) CD-HIT:
    Do step 1 of BPGA using your gbk or protein fasta files. Use the INPUT_all.faa/.seq file generated here as input for cd-hit (Set your own identity and coverage values). You can either use online cd-hit web server or standalone in linux.
    Now use *.clstr or *.sorted files from the cd-hit output in Step 2 of BPGA. (copy the file in bin folder and enter full name when asked). You are done.
    2) OrthoMCL:
    For users new to OrthoMCL get a preinstalled virtual Operating System from this link
    You must run orthomcl on INPUT_all.faa / .seq file generated by Step1 of BPGA. Instead of running orthomcl with separate fasta files, you must start with makeblast db command using -in INPUT_all.faa or .seq file instead of GoodProteins.fasta file.
    BPGA already merged those seperate faa files together into input_all.faa.
    Then use the groups file created by OrthoMCL into Step 2 of BPGA. (copy the file in bin folder and enter full name when asked). You are done.
    (Note that unique genes are not present in groups file, user needs to manually insert unique genes in groups file.)

SUPPORTED INPUTS

BPGA requires any of the three types of input files of your dataset for analysis (*.gbk or Protein FASTA files from NCBI and HMP databases or any protein Fasta or binary 1,0 matrix).

    -NCBI GenBank File

    -NCBI Protein FASTA files sample:

    >gi|19745202|ref|NP_606338.1| protein name [Organism Name]
    MTENEQIFWNRVLELAQSQLKQATYEFFVHDARLLKVD
    MRTNFKVSFYLRSNYENKEGKSPVMLRVFLNGEMSNFG

    -HMP Protein FASTA files sample:

    >HMPREF9420_0006 protein name [Organism Name]
    MRTNFKVSFYLRSNYENKEGKSPVMLRVFLNGEMSNFG
    MTENEQIFWNRVLELAQSQLKQATYEFFVHDARLLKVD

    -Any Protein FASTA files sample:

    >Any_header_information
    MRTNFKVSFYLRSNYENKEGKSPVMLRVFLNGEMSNFG
    MTENEQIFWNRVLELAQSQLKQATYEFFVHDARLLKVD

Note: All gene bank (.gbk) or Protein FASTA files should be concatenated in single file in case of more than one file for a single organism for example more than one chromosomes or contigs or scaffolds.

See BPGA User Manual.pdf for detailed instructions.

Email your queries to : bpgatool[at]gmail.com

How to cite this article:
Chaudhari, Narendrakumar M., Vinod Kumar Gupta, and Chitra Dutta. "BPGA-an ultra-fast pan-genome analysis pipeline." Scientific reports 6, 24373 (2016) ; doi: 10.1038/srep24373.