Getting Personalised Medicine to Market Faster

Professor Jonathan Mullins

Moleculomics Ltd is a Swansea-based SME established by Jonathan Mullins, an Associate Professor at Swansea University’s Institute of Life Sciences, in order to ensure that the outputs of his academic research are used to benefit society and with the aim of creating new wealth and jobs in the Welsh digital life science sector.

Dr Mullins aims to significantly reduce the time it takes for pharmaceutical companies to get their new drugs to market. Today a typical candidate drug compound takes around twelve years to move from idea to market. Dr Mullins’s goal is to significantly reduce this time, potentially saving many lives that could have been lost while waiting for a new cure to come to market.

Moleculomics’ core business (and Dr Mullins academic research) is focused on developing new computational tools (known as pipelines) that take genetic information (in the form of DNA sequences) and automatically convert this into detailed three-dimensional models of all of the proteins within our bodies. Detailed knowledge of every possible structure of a protein would be very valuable to pharmaceutical and biotechnology companies as they move towards personalised medicine – where doctors can identify specific treatment strategies for particular people with particular forms of a disease and tailor clinical management to individual genetic composition.

As an example, they could use the pipeline to carry out simulations of the interactions of a new drug with all of these proteins with a much higher specificity. This would mean that drugs could be made more accurate in targeting the correct protein and would also ensure that they are less likely to result in unforeseen side effects due to interactions with other proteins. As Dr Mullins explains:

“A drug company developing a new medicine might want to screen compounds based on their interaction with a target protein. So, they might see that the drug binds something in the heart – fantastic! But what else does it bind? Could there be side effects? In the past, the drug discovery pipeline has essentially been focused on the interaction between a drug and the cells and proteins of interest. But our new pipeline would enable in silico screening against every protein in the body to identify potential adverse effects. We’d like to see technologies like ours increase the rate of discovery of new drugs and make better and safer drugs that can make a major difference to the lives of people all over the world.“

Access to high performance computing (HPC) is vital in enabling Moleculomics’ pipeline to run fast enough to be useful to a pharmaceutical company. Moleculomics has been working with HPC Wales since May 2012 and this has enabled it to make significant progress in the development of its HPC drug discovery pipeline. According to Dr Mullins:

“Without HPC capacity to be able to compute all of the structures throughout the body and then to screen each of these against the drug compounds we wouldn’t be able to move forward with this technology. For the pharmaceutical industry access to a pipeline like ours could mean that, instead of looking at thousands of compounds with a hit rate of a fraction of a per cent, they can focus all of the wet lab experiments on compounds that they know are more likely to be effective in targeting the correct protein.

Protein Image

“By around 2002, people had begun to develop tools for some rudimentary protein modelling. Not pipelines – this was semi-automated at best. Then it would take three weeks, to model a single protein. By 2008 we had a fully automated pipeline running on our first Linux box and that took the analysis time down to two hours per protein – and we thought we were the bee’s knees! In 2010 we got a much stronger Linux cluster and that enabled us to bring the typical computational time for a protein down to below an hour – probably about 45 minutes. Again, we thought this was brilliant because we could leave tens of proteins running over night. But the real revolution was when we took that same pipeline, optimised it, and put it on HPC Wales. Now we can run about 300 proteins an hour. We can run a complete bacterial genome – around 5,000 proteins – comfortably in a day. So, we’ve gone from 3 weeks per protein in 2002 to 12 seconds in 2014. And, in fact, as we have databases of the structures, from the end user perspective it is 3 seconds from asking a question about a specific protein structure to getting the answer – which is life changing!

As a result of its work with HPC Wales, Moleculomics has been able to steal a march on its competition through its ability to compute structures for all variants of proteins that are present in the population and is beginning to run screening experiments using docking algorithms to investigate the interaction of these variant structures with candidate compounds for drugs. According to Dr Mullins the capabilities of the Moleculomics pipeline are genuinely revolutionary and provide an opportunity to revolutionise the drug development process:

“I don’t think anybody is doing what we are doing on a pan-genome or multi-genome basis, with the speed and power that we have here in Wales. It couldn’t have been done in the past, but now we have an alignment of genome sequences tied in with the capacity to develop algorithms and pipelines along with the massive computational power of HPC Wales. We needed all of those ingredients to be there to develop this ground breaking technology. Time will tell how ground breaking it really is in terms of translating the capability into new drugs and new approaches for developing drugs. That depends on whether people embrace it. If they do, then we’re talking about many more, safer candidate drugs coming to the fore.

“We could also look at improving our understanding of other organisms. We could develop anti-fungal compounds that could be used as agrichemical compounds, to fight fungal infections in plants. These could also have massive economic importance. So we have the potential not just to change medicines for humans, but also develop anti-bacterial and fungicidal agents for the chemical industry. Our technology could be applied throughout the whole biotechnology realm.”

Dr Mullins hopes ultimately, that by making drug development more efficient, the period of exclusivity on drugs could be reduced. “Thinking towards the future, if we could reduce the amount of investment that pharmaceutical companies had to make to get each drug to market through much more targeted experimenting and drug trialing, then that might justify reducing the duration of the exclusive patent on key drugs of importance to public health in the developing world. This combined with a reduction in the discovery pipeline time could make a real difference. Today, there are large populations who are economically marginalised due to no fault of their own and rely on generic drugs that come to market once the exclusive patent expires. It would be really nice to decrease that time, which is currently 22 years or so and get the generic drug into poorer people’s hands sooner. I know that this is a very grand aspiration, but it’s an incremental thing. Even if we took just one year off the time that it took for a generic version of a drug to become available, we would save many lives.”

Life Sciences is a key sector for the Welsh economy, employing roughly 10,000 people and contributing over £1.3 billion each year to the economy. According to the Welsh Government the sector has grown by 19% over the past three years – and Wales accounts for about 10% of employment in the Life Sciences sector in the UK as a whole (compared with 5% of population).

Moleculomics is an example of how HPC Wales can play a vital role in new business creation in this vital sector. Dr Mullins believes that access to state-of-the-art computing facilities has been a key factor in enabling Moleculomics to develop a world-leading bioinformatics pipeline in South Wales, with the potential to revolutionise the way in which drugs are developed and potentially save many lives across the world. He says:

“There are elements of our business that could have been done without HPC Wales, although not with the same ease or completeness. For other parts, however, there would not be a hope without high performance computing. With HPC Wales we have all of the computing that we need in one place. It’s very well supported and very well delivered to the people who need it – and it can be tailored specifically to their needs at the time. Their team has been excellent – a great catalyst for the development of spin-out ventures in Swansea and throughout Wales. It’s fair to say that Moleculomics wouldn’t exist if it wasn’t for access to HPC Wales.”