- De
- En
Life Sciences and Bioinformatics rely on efficient access to computational resources. To empower researchers in this area, the GWDG offers a number of hosted services, license servers, compute clusters, support, and software.
Our service offer is divided into a Basic and Advanced tier. The Basic tier covers all the GWDG stakeholders (users from the University of Göttingen, UMG, and MPG) and provides low-level support for the use and operation of the HPC clusters. The Advanced tier is restricted to stakeholders that directly contribute funding to our HPC and Scientific services, and dedicated projects such as NHR and KISSKI. This includes consulting, on-demand installation and support of specialized HPC software, standalone bioinformatics servers and services, dedicated courses, and more.
Galaxy is an open-source, workflow-based platform, mainly dedicated to genomic analysis but also for broader data-intensive biomedical research. The platform offers the possibility of setting up workflows and working with histories, keeping your data organized and your research reproducible. Workflows, histories, and data can also be shared internally and even across Galaxy instances, enabling collaborative research.
Our hosted version offers the ability to install all the tools you need from one of the central Galaxy repositories or develop them yourself, and large personal storage capacity. The platform is open to all users with a full GWDG account.
The Jupyter-HPC hub provides access to many services, including a Jupyter Notebook (interactive Python code) Server. This runs directly in the interactive queues of our HPC clusters, which provides ample hardware resources (including GPUs), and access to your data if you are already part of a compute project. Please see the documentation for more information. User-generated custom containers can be run on the service, with the following services already available:
RStudio is an IDE (Integrated Development Environment) for the R programming language. With this service it is possible to interactively develop and program scripts with many IDE features.
For applications that work exclusively or primarily in graphical mode, there is the possibility to run them using Linux-based desktop environments. These run directly in your browser, and like the other Jupyter services have access to cluster resources and project data.
RShiny is a framework for developing interactive web applets in the R programming language. These interactive webpages are an ideal companion to publications in an increasingly multimedia world. The server offers the possibility of freely hosting your app with us.
If you want to host your app with us, please get in touch. Currently, apps are set up manually by an admin. Ideally, your app should already be set up on some version control software platform, such as GitLab (GWDG’s GitLab instance) or GitHub, which makes code sharing easier.
Depending on your affiliation, you can access a variety of compute clusters managed by the GWDG, with different requirements and available hardware. Compute clusters provide access to large CPU and GPU resources, as well as storage and RAM (upwards of 1 TB depending on the system), managed by submitting batch jobs to a scheduler. MPG and University of Göttingen/UMG users can work on the Scientific Compute Cluster (SCC), MPG groups can make use of the new MPG GPU Cloud, research groups from any university can apply for a project on the NHR cluster, and finally the KISSKI AI Service Center is open to all with a focus on AI/ML training and inference. Our GWDG Academy provides regular courses on cluster computing in general, and bioinformatics tasks in particular.
For groups with large compute requirements, we can offer housing and hosting services for your own dedicated hardware and compute nodes.
As Machine Learning and AI models get more popular, access to GPUs for both training and inference becomes more crucial for daily research. Most of the compute clusters mentioned here provide access to GPUs. We recommend starting your work by testing the GPUs available in the Scientific Compute Cluster, and then scale up your processing with the resources of the NHR cluster or KISSKI AI Service Center.
If you want to host your own website or light service, we provide access to Cloud (easier to operate) and ESX (more powerful) virtual machines.
In our GWDG Academy, we constantly offer courses related to technical topics. These include periodic beginner courses such as introduction to working on our compute clusters, and advanced specific topics related to life sciences and bioinformatics. Current topics include:
Courses can also be provided upon request as part of the Advanced service tier if a minimum quota of participants can be guaranteed.
Biological research often involves data that needs the highest level of security, due to privacy regulations on medical and human related data or because of proprietary information. For this purpose, we offer the Secure-HPC service, which combines compute and encryption to ensure your data remains safe in-transit and while at rest when being processed in our compute clusters.
We provide the usual genomic and alignment tools such as bwa and bowtie, but also advanced machine learning based inference tools like evo2. We also provide access to genomic databases on the cluster so you don’t have to download these large datasets. More can be provided upon request.
In recent years, the field of protein study and design has seen a lot of success from machine learning tools, for example, Alphafold. We provide access to various protein folding programs on-cluster and are working on expanding this offer into a full computational protein folding and design hub. We also provide access to the Alphafold-2 and Boltz folding algorithms on the browser through the Protein-AI platform. Introductory courses to the topic are offered through our GWDG-Academy.
Life science research often requires complex and repetitive workflows for data processing. Workflow tools provide the means to implement such workflows into a script, making them consistent, organized, repeatable, and shareable. On the cluster we offer Snakemake and Nextflow, and can provide help in their usage. We also periodically provide an introductory course to Snakemake.
We provide many of the open-source or free ab-initio (Quantum Espresso, CP2K, CPMD, and more) and classical (LAMMPS, GROMACS, NAMD) molecular simulation packages. For licensed programs such as VASP, we can usually provide access to cluster-optimized code if you already have acquired your own license. For others, such as Turbomole in NHR, we offer access based on our own license.
Modern microscopy workflows generate large amounts of data that needs to be stored and post-processed with a variety of methods. Our compute services provide ample storage, as well as the processing power (including GPUs) to analyze these results. As concrete examples, we provide access to tools such as Relion, and example containers for napari and noise2void.
Neuromorphic computing is a computing paradigm based on the behavior of biological neuron networks. It can be used for a variety of problems, including simulating complex networks of realistic neurons (theoretical neuroscience), general network simulations, various graph-based optimization problems, live signal processing, and control loops. We host 4 boards of the SpiNNaker-2 neuromorphic computing architecture, which together can simulate up to hundreds of thousands of neurons. We are currently exploring various simulations with these boards, and access can be granted to researchers to use the boards upon request.
We offer access to the statistics packages SAS, SPSS and Stata. SAS and SPSS are available as paid licenses, while a permanent Stata module is available to users of the Scientific Compute Cluster. For more information, please consult the documentation, and contact our licensing team if you wish to acquire one of these licenses.
GWDG also offers several license servers for validation of programs that require a license. For users of the Advanced service tier, we can help setup such servers for licensed software you might have acquired. Please submit a support ticket.
If you need software not currently available on one of the compute clusters, it is possible to request new installations as part of the Advanced service level. But we also offer various options for installing programs on your own. You can read more about this in our HPC documentation.
Please get in touch if you have any questions, comments, or suggestions. Particularly welcome are suggestions, especially from users of the Advanced service tier, for new services or licenses, long-term software or HPC projects that could be developed jointly, as well as tutorials for currently offered services.
