Combating the coronavirus with computational modelling

    30 April 2020

    Professors Paolo Carloni and Giulia Rossetti are molecular modelling specialists in HBP’s research areas covering Simulation and Drug Design.

    © Forschungszentrum Jülich/Sascha Kreklau

    Prof. Paolo Carloni and Jr Prof. Giulia Rossetti

    Institution: HBP partner Forschungszentrum Jülich, Institute for Advanced Simulation and Institute for Neuroscience and Medicine, Germany

    HBP research areas: Molecular Neuroscience

    Professors Paolo Carloni and Giulia Rossetti are molecular modelling specialists in HBP’s research areas covering Simulation and Drug Design. Their supercomputing-powered simulations provide insight on the workings of signaling molecules in the brain and led to patented drug candidates for neurodegenerative diseases. Recently, they have also devoted additional efforts to combat the Coronavirus pandemic.

    Q: What are you doing at the moment to help address the challenges posed by the novel Coronavirus?

    Molecular dynamics simulations predict structure, dynamics and thermodynamics of biomolecules such as proteins and RNA. These can be complemented by fast data science methods to identify the binding poses of drug candidates onto their receptor targets. Our team at the INM-9 and JSC develops and applies these powerful approaches to neuronal proteins and neuro-active drug design. But a protein is a protein, whether it is located in a neuron or it comes from a virus. Hence, we can essentially transfer our technologies and methods effortlessly to this new challenge.

    We are now participating in the project EXSCALATE4CoronaVirus within the framework of the EU coronavirus H2020 program. It involves the Jülich, CINECA, Barcelona Supercomputing centers, the Dompè Pharma company, the Fraunhofer Institute, and others, for a total of 18 computational and experimental groups. The aim of the project is to identify, by virtual screening, effective antiviral drugs against proteins responsible for the virus survival, among the currently commercially available drugs. This so-called process of ‘drug-repurposing’ is a common procedure for new diseases and at times it has allowed fast identification of safe drugs. The supercomputers allow performing virtual screening combined with biochemical and phenotypic high-throughput screenings. This means billions of molecules against selected targets (proteins of the virus) can be evaluated in silico by the end of the project"

    As a next step, together with Prof. Lippert and Prof. Amunts, we plan to offer access to the powerful e-Infrastructures of the Human Brain Project, such as Fenix. Fenix is a unique supercomputing network and the joint effort of five of the largest European supercomputing centers.  The e-Infrastructure  has some unique features that can boost the efforts of the molecular simulation community at large. A distinguishing characteristic is that data repositories and scalable supercomputing systems are in close proximity and really well integrated. In fact, Fenix provides interactive computing, virtual machines, active and archival data repository services all at the same time. From the outset, the community will be supported by the High Level Support Team (HBP-HLST) of HBP and the "Neuroscience" simulation laboratory of the Jülich Supercomputing Centre to have a head start in the use of the services. By opening FENIX to the worldwide molecular simulation community, we hope that a great number of researchers in the field will exploit this powerful infrastructure for their research on the coronavirus.

    Update (28. 04. 2020)

    Webinar: Giulia Rosetti gave first public report on our preliminary results on COVID-19 EU grant in this webinar organized by CECAM (43:30)

    <iframe allowfullscreen="true" frameborder="0" height="410" src="" style="max-width:100%;" width="729"></iframe>

    Update (29. 04. 2020)

    The Fenix Research Infrastructures are being made available to researchers and scientists working on topics related to the current COVID-19 pandemic. Read more