Dr. Guillermo del Angel is currently Sr. Director for Data Science, Genomics and Bioinformatics at Alexion Pharmaceuticals. He has been leading research efforts focused on applying and developing different machine learning, genetics and data science approaches to improve rare disease target discovery, drug development, and patient/disease characterization

FLG: Could you give us an introduction of yourself and the work you do?

I’m Guillermo del Angel and I’m the Senior Director for Data Science Genomics and Bioinformatics at Alexion Pharmaceuticals. We are a rare disease company based in Boston, Massachusetts in the United States. We focus on rare diseases and have four drugs in the market right now with others in the pipeline. We’re very excited about using genomics to help our patients.

 

FLG: What drugs has Alexion Pharmaceuticals developed?

Our largest drug by revenue is still Soliris®, used originally to treat paroxysmal nocturnal haemoglobinuria (PNH) and atypical haemolytic uremic syndrome (aHUS) which are very rare blood disorders. It is now also being used as well to treat myasthenia gravis and neuromyelitis optica which are neurological disorders. These diseases are caused by dysregulation of the complement system, the area that Alexion originally developed deep scientific expertise on. We recently launched Ultomiris®, our next-generation, long-acting complement inhibitor for the treatment of PNH and aHUS, with several other indications in the pipeline.

We also have two metabolic drugs – Strensiq® for hypophosphatasia and Kanuma® for lysosomal acid lipase deficiency. They are all very rare diseases, but very devastating, and we’re hoping to make a big impact on our patient’s lives.

 

FLG: What do you think is important during drug development from the research to treatment?

It’s a very long process. Developing a drug is not easy and it takes many years. I think the main thing we need to have is resilience and tolerance for failure because unfortunately, a lot of things we think will work eventually do not. You need to have the mindset of always trying to do the best for the patients, and if something fails, learn from the experience and move on. So, its not just the scientific skills that you need to acquire, but also the resilience and learning from failure that you need to be able to succeed.

 

FLG: Why are we seeing an increase in the demand for genetic data?

I think people are starting to realise more and more the value of that data. Understanding the genetic data and making sense of genetic data is a huge first step in developing new treatments and healthcare. It’s also driven by the fact that we can now sequence genomes in a way that is much cheaper and easier than before.

 

FLG: What do you think the future of machine learning and artificial intelligence holds?

I think the future is very bright. Image processing is a perfect case where we have seen AI outperforming human beings for its precision and its ability to make sense of images. More AI is now being applied to making sense of electronic medical records, large volumes of data that can be difficult to analyse and make sense of. But people are realising now that you can use all these new AI and machine learning approaches to make sense of that, an example being to identify patients that have undiagnosed problems through pattern recognition and artificial intelligence.

 

FLG: How can research cover the needs of patients through treatments that do not offer economic profit in the pharmaceutical industry?

We have seen a lot of public entities and governments support basic research and support that translational part for situations in which it might not be economically feasible for a private entity to do that. So, I think it’s very exciting to see all these novel partnerships and all of these innovating funding mechanisms that are brought to life to solve this problem.

 

FLG: What breakthroughs do you see happening in genomics?

The main obstacle right now with genomic data to have an impact is making sense of it and interpreting it. It’s no longer about the cost of the sequencing, but the interpretation. There’s a lot of work going on with how to interpret variants and how to make sense of genomic data, and I think that soon we’re going to keep on making progress on that.

 

Interested in Data-Driven Drug Development? Check out our D4 Europe event: https://www.d4-europe.com/