Digitalising drug discovery

The oceans of health data out there can be overwhelming for pharma companies to manage – but if extracted correctly, the prospect to develop drugs from scratch in as little as a year is very real, says Lifebit CEO, Dr Maria Chatzou Dunford.

As data and digital technology become vital to every aspect of life sciences, the industry is increasingly looking beyond biologists, chemists, and doctors to drive its drug development – and finding that technology has a chief role to play in the future of medicine.

According to an article by Stephens, Zachary D., et al. on Big data: astronomical or genomical? by 2025 more than 500 million human genomes will be sequenced, creating more data than YouTube and Twitter combined. Mining this data to advance drug discovery and new scientific breakthroughs relies on overcoming the overwhelming conundrum of extracting meaningful insights from massive data that is distributed, non-standardised, complex, and inaccessible to most.

Dr Maria Chatzou Dunford, a bioinformatician by background, recognised the fundamental role technology could play in accelerating drug discovery through overcoming these challenges, and as a result founded AI-bioinformatics technology company Lifebit in 2017.

Lifebit’s mission is to revolutionise bioinformatics and biomedical data analysis by bringing together biobanks from across the globe to create an “access portal to the world’s clinico-genomic data,” Dunford says. The company’s Lifebit CloudOS platform enables researchers to query, analyse, and collaborate across large distributed sets of genomic and medical data regardless of where it resides.

The idea to found the company came when Dunford and her co-founder Dr Pablo Prieto Barja were themselves working on analysing genomic and biomedical data for research purposes.

Bioinformaticians by trade, they felt the pain of analysing this data firsthand.

“We found ourselves spending 90% of our time dealing with computational data hassles rather than focusing on the biology and the results,” Dunford says. “Gradually we realised this problem was becoming a norm for the entire industry, and that’s when we founded Lifebit.”

Dunford also believes the industry has just entered a new ‘Genomics 2.0’ era. Legacy technologies are built for an old genomics model – a world with very few, very small centralised genomic datasets that were not very diverse.

“Today, companies have exponentially more datasets, and genomic data by itself is no longer enough. They need clinical, phenotypic, and observational data to supplement genomic data to uncover next-level insights,” says Dunford.

“There’s added complexity in that all this data resides across multiple different sites – including research institutions, clinical settings, pharma companies and biotech companies.”

The ability to bring all this data together, she says, will completely change drug discovery and give companies an important competitive edge.

“If you look at the history of pharmaceuticals, initially it was all about chemistry, and it took pharma about 100 years to get that right, and it took another 50 years to start getting biology right.

“The next ‘big thing’ for the new generation of pharma to get right is its approach to data. The industry needs to shift towards operationalising personalised medicine, creating drugs that are more valuable and precise, and unlocking value-based pricing. But they don’t have 50 years – to stay competitive they need to innovate over the next five years, and investing in Genomics 2.0 technologies could be a game-changer in bringing new drugs to market in just a few years.”

Growing pains with data

Pharma has often failed to keep up with the rapid advances in technology and data. Dunford notes that the amount of data we have today would have been unimaginable even three years ago, and this means that most legacy data platforms that exist within organisations are not built to cope with it.

“There are some innovative pharma companies out there such as Roche and AstraZeneca, but the industry at large is still light years away from harnessing technology to derive data insights to digitalise drug discovery.”

One symptom of this, Chatzou-Dunford says, is companies’ tendency to hire reactively.

Dunford feels the lack of technological advancement in this area is largely due to the fact that historically there have been few data platforms options available, forcing the industry to adopt niche and specialised systems. Consequently, pharma companies hired experts to build in-house systems versus investing in best-of-breed technology.

Accelerating drug discovery

To digitise the drug discovery process, pharmaceutical companies need to better access and manage data, but the industry is far from where it needs to be.

“Companies should aim to get their drug discovery to a point where approximately 80% is digital and only 20% physical, with the latter part just being confirmation,” says Dunford. “Right now it’s the opposite – 80% is physical and observational, and sometimes anecdotal, even. That makes extrapolation difficult, and increases the chance of the trial failing.”

Access to population data speeds the process as the majority of experiments needed to develop some drugs have already been completed in the real world.

“Rather than starting with a random drug in the hope that it will treat a particular disease, you can flip drug discovery on its head and look at which patients are more prone to the disease, understand the genetics and protein-functions behind it, and then work backwards to find a chemical to treat it.”

Consequently, drug discovery timelines could be reduced to as little as one year. COVID-19 vaccines have demonstrated pharma R&D’s ability to move with speed, and Dunford sees no reason why similar timelines can’t be achieved for personalised medicines.

“If you have enough data from hospitals across the world, you essentially have pre-existing clinical trial data that you can analyse endlessly, as well as being able to call in those patients for more samples. It brings the clinical trial into the real world.”

Examples of this already exist. Genomics England, for instance, is currently analysing the genetic code of 35,000 patients with COVID-19 to help scientists understand whether a person’s genetics may influence their susceptibility to the virus. A paper published in Nature on Genetic mechanisms of critical illness in COVID-19 has already revealed the genes that are linked to COVID-19 susceptibility.

“We wouldn’t need to lockdown an entire city or country if we knew more about the genomics of COVID-19,” says Dunford. “Instead, we’d only need a specific group of people to stay indoors.”

The industry could also start to take a disease-wide approach to drug development by selecting a disease, gathering all the related population and clinical data, bringing together the right tools and experts to analyse and assess potential treatments, and then manufacture the right pill.

“First, though, the entire industry needs to get more data into a state where it can actually be used for better understanding of the underlying disease-genetics, diagnosis, prevention, treatment, and drug discovery.”

About the interviewee

Dr Maria Chatzou Dunford is the CEO and co-founder of Lifebit.AI. Maria is a thought-leader and biotech innovator, expert in AI-driven drug discovery, biomedical informatics and federated computing. She is also a passionate entrepreneur and has founded two companies, Innovation Forum Barcelona and Techstars-backed Lifebit. Prior to Lifebit, she was a biomedical researcher, working on developing tools and methods that facilitate the analysis of Big Biomedical Data and promote personalised medicine discoveries. This includes the industry’s standard programming framework, Nextflow.

About the author


George Underwood is a senior member of the pharmaphorum editorial team, having previously worked at PharmaTimes and prior to this at Pharmafocus. He is a trained journalist, with a degree from Bournemouth University and current specialisms that include R&D, digital and M&A.

Don’t miss your complimentary subscription to Deep Dive and our newsletter

Sign up



Your name

Your e-mail

Name receiver

E-mail address receiver

Your message