Cando is an open-source, computational chemistry environment for building and designing functional molecules and materials.
- apt-get for Ubuntu 22.04, Debian Bookworm (in testing) and Windows Subsystem for Linux (WSL2).
- homebrew for MacOS.
- pacman for Arch Linux.
These installations can be automatically updated from the Nighly Package Repository for each operating system to get bug fixes and the most up-to-date version of Cando. We are fixing (and creating :-)) bugs all the time.
If you want to use the jupyter-lab kernel implemented in Cando then follow the entire instructions below for each operating system. If you only want to install Cando - then skip the steps that mention jupyter-lab.
Binary packages for Cando are built nightly starting about 2:00 AM EST and take about 4 hours to complete. Currently packages are built for Ubuntu 22.04, Debian Bookworm (in testing), Arch, and homebrew for macOS. The Ubuntu 22.04 package works on WSL2 (Windows Subsystem for Linux). To add the package repository to your local system and install Cando execute one of the following commands in a shell. You may need to prefix these commands with sudo for root access.
-
Install jupyter-lab using miniconda. This can be skipped for distributions with very new JupyterLab versions like Arch.
conda install -c conda-forge jupyterlab nodejs=17.9.0 sidecar
-
Install Cando (this should detect and work for both Ubuntu or Debian)
sudo bash -c "$(curl -fsSL https://www.thirdlaw.tech/pkg/cando.sh)" -
Install Cando user snapshot and Jupyter kernels
cando-user-install
Now the installation is done! Scroll to the instructions below to download cando-demos and run jupyter-lab/cando!
If step 2 fails to detect your operating system then you can install manually via one of the following.
- Debian Bookworm
curl -fsSL https://www.thirdlaw.tech/pkg/key.gpg | gpg --yes --dearmor --output /etc/apt/trusted.gpg.d/thirdlaw.gpg add-apt-repository "deb https://www.thirdlaw.tech/pkg/bookworm/ ./" apt-get update apt-get install cando
- Ubuntu 22.04 (Jammy)
curl -fsSL https://www.thirdlaw.tech/pkg/key.gpg | gpg --yes --dearmor --output /etc/apt/trusted.gpg.d/thirdlaw.gpg add-apt-repository "deb https://www.thirdlaw.tech/pkg/jammy/ ./" apt-get update apt-get install cando
- Arch
curl -fsSL https://www.thirdlaw.tech/pkg/key.gpg | pacman-key --add - pacman-key --lsign-key 009C1F20737D81230A0928CBC812E10EC39DF202 echo -e "\n[thirdlaw]\nServer = https://www.thirdlaw.tech/pkg/arch\n" >>/etc/pacman.conf pacman -Sy --noconfirm cando-git
Binary packages for MacOS Big Sur and Monterey are available from the clasp-developers/clasp tap.
Currently Cando only runs on x86-64 systems or Apple Silicon using rosetta2. The rosetta2 environment on Apple Silicon is excellent and quite performant. In the near future we will support native Apple Silicon - this is awaiting updates from Apple to the llvm JIT compiler.
To install Cando do the following:
-
Install homebrew for x86.
If you have an x86 Mac then you can proceed. If you have an M1/Apple Silicon Mac then set up an x86 shell and open that shell.
Homebrew installed from an x86 shell installs in
/usr/local. -
Install Cando
brew install clasp-developers/clasp/cando
-
Install Cando user snapshot and Jupyter kernels
cando-user-install
These Homebrew formula can be built from source by adding the --build-from-source option and can be built directly the most recent commit by adding the --HEAD option.
After the initial install, to upgrade to the latest nightly build use brew update && brew upgrade. The brew update only needs to be done occasionally to update homebrew itself.
Several JupyterLab notebooks that demonstrate how Cando can be used for computational chemistry can be found in Cando-demos.
To run jupyter-lab and start a demo:
- Clone the cando-demos.
git clone https://github.com/cando-developers/cando-demos.git - From your home directory run
jupyter-lab - Copy and paste the provided link into a web browser (Firefox is recommended).
- Double click on the filename "00-Welcome-and-README.ipynb" and it will open the jupyter-lab notebook.
For operating systems where jupyter-lab cannot launch a browser like remote linux systems and
WSL2 (Windows Subsystem for Linux) you should use jupyter-lab --no-browser and then copy
the entire URL generated by jupyter-lab into your browser.
For up to date video demos and tutorials please subscribe and like our videos on YouTube.
Thermodynamic integration using a pre-built ligand library
Sketching molecules for thermodynamic integration calculations