deMEM: a novel divide-and-conquer framework based on de Bruijn graph for scalable multiple sequence alignment

deMEM is a program written in C++ for aligning MSA based on de Bruijn graph with divide-and-conquer framework. It runs on Linux and Windows.

Usage

        Usage: samemalign_main (64-bit) [options]

Use deBruijn graph find MEM, and make alignment by a specified aligner like WMSA/MAFFT/HAlign 3
Available options:
File process:
        --in         FILE      input sequence file name (Required)
        --out        FILE      output file name (Required)
        --removechr            remove characters in sequence
Config JSON file: (Recommended)
        --json       FILE      JSON config, contain "Conditions", "MEM arugments" and "Alignment arugments" configurations
        --jsonsample FILE      output JSON config sample to FILE and exit
        --keepjson             keep JSON file when finish
Conditions:
        --thread     N         use N threads to run program (default: 1)
        --type       X         sequence type: X = 1 is Protein, X = 2 is DNA (Required)
MEM arugments:
        --threshold  t         the minimal length of MEM (default = 10)
        --memregen             re-generate MEM (default: false)
Alignment arugments:
        --highsim              this file is high similarity, will treat all sequences into one cluster; will ignore --cluster,
                               --wmsac and --mafftc
        --path       PATH      temp file will be in path, default = "swap/" (only support first-level directory)
        --threshold2 t         the minimal length of blockaligner (default = 10)
                               not recommended for t < 10
        --align      A         alignment method: A = 0 is WMSA, A = 1 is MAFFT, A = 2 is HAlign, A = 3 is abPOA (default: 0)
        --wmsaa      w         WMSA align method: w = 0 use FFT+K-band align, w = 1 only use K-band align (default: 0)
        --maffta     m         MAFFT align method: m = 0 is NW-NS-1, m = 1 is NW-NS-2, m = 2 is NW-NS-i, m = 3 is FFT-NS-1,
                                                   m = 4 is FFT-NS-2, m = 5 is FFT-NS-i, m = 6 is G-INS-i, m = 7 is L-INS-i
                                                   (default: 3)
        --gapopena   a         use user-defined gap open/gap penalty in align for WMSA, MAFFT and abPOA
        --gapexta    a         use user-defined gap extension in align for MAFFT and abPOA
        --merge      M         merge two profiles method: M = 0 is WMSA, M = 1 is MAFFT, M = 2 is abPOA (default: 0)
        --wmsam      w         WMSA merge method: w = 0 use FFT+K-band with no gap open calculation,
                                                  w = 1 use K-band with no gap open calculation,
                                                  w = 2 use FFT+K-band with gap open calculation,
                                                  w = 3 use K-band with gap open calculation (default: 0)
        --mafftm     m         MAFFT two profiles align method: m = 0 is NW-NS, m = 1 is FFT-NS (default: 1)
        --gapopenm   m         use user-defined gap open/gap penalty in merge two profiles for WMSA, MAFFT and abPOA
        --gapextm    m         use user-defined gap extension in merge two profiles in WMSA, MAFFT and abPOA
        --cluster    C         merge clusters method: C = 0 is WMSA, C = 1 is MAFFT, C = 2 is abPOA (default: 0)
        --wmsac      w         WMSA cluster merge method: w = 0 use FFT+K-band with no gap open calculation,
                                                          w = 1 use K-band with no gap open calculation,
                                                          w = 2 use FFT+K-band with gap open calculation,
                                                          w = 3 use K-band with gap open calculation (default: 1)
        --mafftc     m         MAFFT cluster merge method: m = 0 is NW-NS-1, m = 1 is NW-NS-2, m = 2 is NW-NS-i, m = 3 is FFT-NS-1,
                                                           m = 4 is FFT-NS-2, m = 5 is FFT-NS-i, m = 6 is G-INS-i, m = 7 is L-INS-i
                                                           (default: 3)
        --abpoac     p         abPOA cluster merge method: p = 0 use center sequences make pre-align and insert clusters,
                                                           p = 1 directly align clusters (default: 0)
        --gapopenc   c         use user-defined gap open/gap penalty in merge clusters for WMSA, MAFFT and abPOA
        --gapextc    c         use user-defined gap extension in merge clusters for WMSA, MAFFT and abPOA
        --allmafft             all method use MAFFT
        --allwmsa              all method use WMSA
        --allabpoa             all method use abPOA
Others:
        --noverbose            not print the info for subprograms
        --help                 print help message
        --version              show program version
Hint: If read json file and use same arugments, program will omit the arugment settings.
Example:
        samemalign_main --in seq.fasta --out seq_out.fasta

We strongly recommended run the program by determined json files, which in config folder. You can test by this command:

# Install by the following sections 
samemalign_main --json config/wmsa.json --in test/seq_1.fasta --out test/seq_1.ans.fasta

Installation

Linux/WSL (Windows Subsystem for Linux) - from Anaconda (Recommended)

1.Install WSL for Windows. Instructional video 1 or 2 (Copyright belongs to the original work).

2.Download and install Anaconda. Download Anaconda for different systems here. Instructional video of anaconda installation 1 or 2 (Copyright belongs to the original work).

3.Install deMEM.

#1 Create and activate a conda environment for deMEM
conda create -n demem
conda activate demem

#2 Add channels to conda
conda config --add channels wym6912
conda config --add channels malab
conda config --add channels conda-forge

#3 Install deMEM
conda install -c malab -c conda-forge -c wym6912 demem

#4 Test deMEM
samemalign_main --help

Linux/WSL (Windows Subsystem for Linux) - from the source code

1. Download and Compile the source code. (Make sure your version of gcc >= 11 or clang >= 13.0.0)

#1 Download
git clone https://github.com/malabz/demem.git --recursive
# Make sure submodules are correctly cloned
git submodule init
git submodule update
git submodule foreach git submodule init
git submodule foreach git submodule update

#2 Open the folder
cd demem/src

#3 Compile and install to path
make THREADS=16 PREFIX=/path/to/you/wish all

#4 Test
./samemalign_main --help

Windows - from from Anaconda (Recommended)

1.Download and install Anaconda. Download Anaconda for different systems here. Instructional video of anaconda installation 1 or 2 (Copyright belongs to the original work).

2.Install deMEM.

#1 Create and activate a conda environment for deMEM
conda create -n demem
conda activate demem

#2 Add channels to conda
conda config --add channels wym6912
conda config --add channels malab
conda config --add channels conda-forge

#3 Install deMEM
conda install -c malab -c conda-forge -c wym6912 demem

#4 Test deMEM
samemalign_main -h

Windows - from Visual Studio 2022 (Not Recommended)

1. Firstly, install git, and download the whole project with submodules:

git clone https://github.com/malabz/demem.git --recursive
# Make sure submodules are correctly cloned
git submodule init
git submodule update
git submodule foreach git submodule init
git submodule foreach git submodule update

2. Make programs from Visual Studio 2022:

• First of all, download and install Visual Studio 2022.
• Open memalign.sln in Visual Studio 2022, then switch it to Release mode (Select Properties, choose Configuration Properties, then press Configuration Manager... button, change Active Solution configuration to Release), choose Build -> Build Solution. If failed, please try compile project one by one.
• Open x64\Release folder, copy all the .exe and .dll files to x64\samemalign_env\bin.
• Copy src\external\submafft\windows\{profiles_wrapper.exe,src} file/folder to x64\samemalign_env.

Dataset

The data used in this article is in zenodo.

Citation

Contacts

If you find any bug, welcome to contact us on the issues page or email us.

More tools and infomation can visit our github.