We have written an article to publish the method for network embedding able to leverage multiplex heterogeneous networks and this was submitted in January 2020. We are also finalizing the code, which will be available through github. Overall, the short-term collaboration project has allowed us to finalize and evaluate in-depth the first algorithm, multiplex network embedding and open the path to the second, which can leverage multiplex-heterogeneous networks in the network embedding task. The code is finalized, we have to run the simulations. It is, to our knowledge, the first one that can deal with this kind of network. This is particularly important as it opens the toolbox of machine learning for network data. This method will be useful not only for the Computational Biology community, but also for Computer Science and Physics communities working with large-scale and complex networks.

It is to note that I also appreciated the general activities of the lab, including the group meetings, discussions, the presentation of my work and results. This allowed me to create a strong network of collaborations.

Jointly with NTUA, I have done considerable work fitting biologically realistic neuron models (here: Hodgkin-Huxley extensions) to in-vivo data with very high accuracy. By capitalising on the large expertise of the host institution with Intel-Phis, we have jointly finished the following activitis: i) Deployed brain stimulations on an 8-node Xeon-Phi (Knight's Landing, KNL) configuration on the cloud ii) Evaluated the system performance, scalability with respect to netowrk parameters (size, distributiuon) iii) Planned one (perhaps two) publication(s) to a high-impact journal iv) Had discussions on incorporating such promising results to upcoming Cloud-computing grants.

The research performed during the short term collaboration has resulted in the confirmation of our expectations regarding the cost of short-range branches versus long-range branches. The optimizations to the MaxineVM has been pushed on the upstream repository, allowing the broader community of managed runtime systems to benefit from our efforts. In addition to the work done on the direct topic of this grant, the grantee and the host also had the chance to collaborate on another work which resulted in a paper submission to the 12th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage ‘20).

The coronavirus responsible for the current Covid-19 pandemic, SARS-CoV-2, has on it’s surface a protein, called spike protein, that is responsible for the attachment to host cell receptors. This protein is also present in other types of coronaviruses, like SARS-CoV and MERS-CoV, which were responsible for epidemics in the past. In the context of the current Covid-19 pandemic, BSC’s Computational Biology Group is applying computational approaches to better understand what specific regions on the spike protein are important for host specificity, i.e. why some coronaviruses infect human and other coronaviruses infect other kinds of animals. As a part of the Exscalate4CoV project, our group identified important amino acid residues on the SARS-CoV-2 spike protein using a computational analysis based on a multiple sequence alignment. The identified residues were validated using available theoretical/experimental data and will be very important for guiding other types of computational research, like Molecular Dynamics simulations, as well as experimental assays for the development of treatments and vaccines.

The BSC-CNS is collaborating in the fight against the coronavirus in different fronts: 1) Investigating about the virus and possible treatments, by analyzing mutations on the coronavirus genome, and searching for possible drugs and immunologic therapies (antibodies and vaccines); 2) Employing artificial intelligence and natural language processing to analyze data about the propagation and impact of the epidemic in the society, generating data to help decision making; 3) Using the supercomputer MareNostrum 4 to internal research and giving access to external investigators working on the epidemic.