Code repository for the VIB Hackathon June 2024 on spatial omics. More info: https://hackmd.io/@berombau/BJetSxw8T.

See instructions at https://github.com/biohackrxiv/bhxiv-gen-pdf.

Update the paper.pdf using the paper.md.

docker run --rm -it -v $(pwd):/work -w /work biohackrxiv/gen-pdf:local gen-pdf . "Other"

Login to the hpc:

ssh {YOUR_USERNAME}@tier1.hpc.ugent.be

Set the following environment variables in ~/.bashrc:

export SLURM_ACCOUNT='starting_2024_011'
export SALLOC_ACCOUNT=$SLURM_ACCOUNT
export SBATCH_ACCOUNT=$SLURM_ACCOUNT

Start an interactive session, e.g.:

qsub -I -l nodes=1:ppn=16,mem=4G

go to VSC_DATA_VO_USER if a VO is available, else go to VSC_DATA_USER

cd $VSC_DATA_USER

Clone the repository

git clone https://github.com/saeyslab/VIB_Hackathon_June_2024.git

Install minconda there:

mkdir -p miniconda3
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda3/miniconda.sh
bash miniconda3/miniconda.sh -b -u -p miniconda3
rm -rf miniconda3/miniconda.sh
miniconda3/bin/conda init bash

Set libmamba as the solver:

conda install -n base conda-libmamba-solver
conda config --set solver libmamba

Create the environment:

cd $VSC_DATA_USER/VIB_Hackathon_June_2024
conda env create -f env_python.yml

Activate the environment:

conda activate hackathon_multi_omics_2024

Try running the example script:

python examples/read_spatialdata_from_s3bucket.py

This script should create the file test.png in the current working directory.

We also include a simple .pbs script, that can be submitted to the job queue, e.g.:

module swap cluster/dodrio/cpu_rome
qsub examples/run_hackathon.pbs

This will also create the file test.png in the current working directory.

We provide an example on how to launch jobs on the HPC using hydra and hydra-submitit. After cloning the repository and installing the conda environment (see previous section) do the following to launch a job:

module swap cluster/dodrio/cpu_rome
cd $VSC_DATA_USER/VIB_Hackathon_June_2024/examples
chmod +x run_hydra.sh
./run_hydra.sh

A job will be queued, and in the folder examples a directory multirun will be created, containing the logs of your job.

You can play with this example locally if you change the following section of default.yaml from this

defaults:
  - override /hydra/launcher: submitit_slurm

to this

defaults:
  - override /hydra/launcher: submitit_local

However, local testing of hydra-submitit will not work on Windows.

For more complex examples on how to work with hydra, we refer to here and here.

- Open VS Code.

- Go to the Extensions view by clicking on the Extensions icon in the Activity Bar on the side of the window or pressing Ctrl+Shift+X.

- Search for "Remote - SSH" and install it.

- Press F1 to open the Command Palette in VS Code.

- Type Remote-SSH: Open Configuration File... and select it.

- Choose the SSH configuration file to edit. It’s usually located at ~/.ssh/config.

- Add the following configuration to the file:

Host hpc_tier1
    HostName tier1.hpc.ugent.be
    User {YOUR_USERNAME}
    ForwardAgent yes

- Press F1 to open the Command Palette.

- Type Remote-SSH: Connect to Host... and select it.

- You should see hpc_tier1 in the list. Select it.

- VS Code will open a new window and start connecting to the remote server.

- You might be prompted for your SSH key passphrase or password.