Skip to content

Installation

Ricky edited this page Apr 11, 2023 · 12 revisions

Overall

PanGraphViewer is written in python3 and can run on Windows, macOS and Linux systems.

  • We have fully tested the program in Windows 10, Ubuntu 18.04, Ubuntu 20.04, macOS Big Sur and macOS Monterey.

  • For other equivalent systems, PanGraphViewer may work, but for older versions, there might be some problems in installing some python3 libraries, for instance the PyQtWebEngine library.

  • There are two PanGraphViewer versions: Desktop-based version and Web browser-based version.

  • To make it easy to run PanGraphViewer, we strongly encourage users to use miniconda3 to install all dependencies.

  • The minimum python3 recommended is python3.6.

Miniconda3

Depending on the system, users can download miniconda3 at the offical website

To make it easy, we recommond installing pkg version on macOS and exe on Windows.

After the installation of miniconda3, users may follow the steps below to install PanGraphViewer

Preparation for insallation

  1. Open powershell or terminal

    On Windows

    Users need to find or search Anaconda Powershell Prompt (miniconda3) and then open it.

    On Linux/macOS

    Users need to find or search Terminal and then open it.

  2. [Optional] Users can set a PanGraphViewer environment by using

    conda create -n PanGraphViewer
    

    After typing y, users can activate the PanGraphViewer env by using

    conda activate PanGraphViewer
    

Desktop-based version

  1. Users can copy and paste the command line below to install the python3 libraries in the opened terminal or powershell

    conda config --add channels conda-forge
    conda config --add channels bioconda
    conda install pyqt pyqtwebengine configparser pandas bokeh==2.2.3 dna_features_viewer natsort networkx jinja2==3.0
    
  2. One more installation

    On Windows

    conda install m2-base pyfaidx 
    

    On Linux/macOS

    conda install pysam 
    

Web browser-based version

  1. Users can copy and paste the command line below to install the python3 libraries in the opened terminal or powershell

    conda config --add channels conda-forge
    conda config --add channels bioconda
    conda install pip django==3.1.3 django-bootstrap4 requests django-environ djangorestframework configparser natsort networkx bokeh==2.2.3 dna_features_viewer pandas whitenoise jinja2==3.0
    pip install fontawesome-free==5.15
    
  2. One more installation

    On Windows

    conda install m2-base pyfaidx 
    

    On Linux/macOS

    conda install pysam 
    

Alternative installation method

If users prefer using pip to install the python3 libraries, we have provided the option, but it is not recommneded as conflicts may occur during installation. Advanced users could use this option to install the python libraries.

  • If pip is not installed in the conda environment, users may install it using

    conda install python pip 
    

Once pip is ready, users can follow the steps below to install the python libraries for either PanGraphViewer versions

  1. Move to the PanGraphViewerXXX folder, for example the PanGraphViewerApp folder

    On Windows

    If the PanGraphViewer package is in the Desktop folder

    cd C:\Users\%USERNAME%\Desktop\panGraphViewer\panGraphViewerApp
    

    If users have put panGraphViewer on another drive, users may need to move to the target drive first. For instance, the target drive is D, users can move to the drive by typing D: in Anaconda Prompt (miniconda3) and press Enterfirst and then move to the panGraphViewerApp directory by typing

    D:
    cd panGraphViewer\panGraphViewerApp
    

    On Linux/macOS

    If the PanGraphViewer package is in the Desktop folder

    cd ~/Desktop/panGraphViewer/panGraphViewerApp  
    

    NOTE: please give the correct panGraphViewerXXX path

  2. Install the libraries

pip install -r requirements.txt           ## On Linux or macOS systems
pip install -r requirements_windows.txt   ## On Windows systems

If users using Windows, two more libraries are needed and they have to be installed using conda

conda install m2-base pyfaidx