Humanity has an innate desire to understand the world around it and, where possible, adapt nature to benefit humankind. When anthropologists first entered the cave of Lascaux in France in the 1940s, they immediately understood the ancient paintings they saw. Thousands of years earlier, their ancestors had modelled the movements of the animals they needed to survive.

The basic human need to understand, model and copy nature became one of the foundational drivers of scientific discovery. When Leonardo da Vinci drew the Vitruvian Man in the 15^th century, seeking to accurately represent the proportions of the human body, he did so to try and comprehend the nature of the body's physical workings.

Understanding nature through molecular modelling has that same intention. With the discovery of chemical structure, scientists provided models of the first molecules. In 1865, at the Royal Institution in London, German organic chemist August Wilhelm von Hofmann displayed the first molecular models of methane, ethane and methyl chloride, assigning a system of colours to specific elements which remains in use today.

Arguably the most famous example of such modelling came in 1953 when Francis Crick and James Watson presented the first 3D model of DNA. They did not achieve this in isolation, adding to the work of peers such as Rosalind Franklin and Maurice Wilkins.

And it is this spirit of collaboration —building on the discoveries and attempts to understand nature that came before— that has brought molecular modelling into the modern age. Specifically, molecular modelling is now seen as a potential tool in the fight against COVID-19.

Exscalate4CoV, a new consortium of 18 universities, research centers and private corporations across Europe, was created to find a cure to COVID-19 through molecular modelling.

Exscalate4CoV has been supported by the EU's Horizon 2020 programme for research and innovation. At the time of its launch, Thierry Breton, European Commissioner for the Internal Market, described the project as demonstrating “the value of true pan-European cooperation by joining the best capacities Europe has to offer in the fields of biomedical science and high-performance computing… to fight the coronavirus."

At the core of this initiative is the most powerful industrial high-performance computer (HPC) in the world—Eni's HPC5, which has a computing power of over 50 petaflops. HPC5 is carrying out an advanced simulation on molecular dynamic results from studying coronavirus proteins and a database of approximately 70,000 known drugs to see if any of them they could potentially block the virus' activity.

It might seem unconventional for a global energy company like Eni to lend so much computing power to such an ambitious endeavour in the healthcare industry. When HPC5 was first unveiled in February 2020, there were any number of ideas presented to Eni about how its prodigious processing power could be used.

For Alberto Delbianco, Eni's SVP for Downstream R&D who oversees the Green Data Center housing the HPC-5, the project made perfect sense.

“The Exscalate4CoV project was the first and most clearly outlined request that came to Eni,“ says Delbianco. “We contribute to the project in partnership with Cineca, a non-profit research consortium that involves the collaboration of universities, national research centers and the Italian Ministry of Education, University and Research.”

And while Eni's participation is part of a larger European initiative to create treatment solutions for the coronavirus, the results will be made widely available. “It is Eni's concern that the results of the research will be spread as widely as possible, and the nature of the project offers this possibility," explains Delbianco.

Results did not take long. Shortly after the HPC5 supercomputer began simulating the behaviour of known proteins of COVID-19, a potential treatment emerged. In June 2020, the Exscalate4CoV team announced that Raloxifene, a generic drug primarily used to treat osteoporosis, might help treat COVID-19 in patients with mild symptomatic infections. According to the team, the drug had shown it might be effective “in blocking the replication of the virus in cells, and could thus hold up the progression of the disease."
By June 2020, HPC5 and other three supercomputers working on the project had conducted molecular modelling tests on more than 400,000 molecules, covering both man-made drugs and natural products. In the next phase, HPC5 will continue testing existing molecules, but will also try to find specific novel molecules that could provide additional options to combat COVID-19.

Francesco Figerio, Eni's project manager for the Exscalate4CoV project and a molecular modelling expert, within his team, was able to make a quick transition to the understanding of this healthcare challenge. “For over thirty years with Eni, I have covered many topics in materials modelling —from protein engineering, photovoltaic technology, detergents and lubricants activity, and enhanced oil recovery with organic polymers and smart water formulations. All this could also be applied to the Exscalate simulation scheme of viral proteins," he says.

The first screening of a database of known drugs produced a wide range of potential molecular targets blocking viral activity within the simulation. All these molecules were and are being tested for activity and tolerability.

“Among them, Raloxifene is the first best result evidenced by the ongoing multistage testing process," says Figerio. The European Medicines Agency is now due to study the results of Exscalate4CoV and then begin clinical assessments for Raloxifene for potential use against COVID-19. Eni and the Exscalate4CoV consortium believe this will only be the first of a range of treatment solutions identified with molecular modelling.