AI has exploded in popularity throughout the pharmaceutical industry during the last decade, with many of these self-learning algorithms being used in a range of applications within the drug development pipeline. With the growing capabilities of the technology, more companies are beginning to consider utilising AI to optimise their drug development schemes, which previously relied on intensive research and curation by R&D staff.
Annabel Kartal-Allen talks to Francesc Fernández, Senior Director of Data Science at Almirall, about how the company makes use of AI to improve its productivity, speed and success in drug development.
Quality data is just one click away
Previously, curating data in pharma involved hours of staff labour, requiring them to scour the Internet for supporting information. Francesc and his team at Almirall take a different approach: “Drug development is a complex process. A critical activity in many of these steps is extracting relevant information from a wide variety of sources, such as internal documents, publications, databases, images, etc.”
“With a unified platform, we focus on accelerating the discovery of new therapeutic targets by extracting applicable, quality data from a variety of places. The goal is to discover synthesisable molecules — which will lead to new technologies — as well as prioritising drug discovery based on novelty and commercialisation potential prior to validation.”
Who is Almirall?
Almirall is a global pharmaceutical company that specialise in medical dermatology, finding novel treatments for common skin conditions via their R&D programme. The company has utilised AI in multiple facets of its drug development process and will continue to develop its software capabilities alongside Microsoft to further optimise operations.
Francesc Fernández, Senior Director of Data Science at Almirall
Shorter development times are cost-effective
Not only does AI facilitate the successful curation of relevant data, but it also optimises the drug development pipeline by allowing users to follow trends of therapeutic targets and novel molecules.
This can help researchers to discover beneficial molecules faster, rather than relying on trial and error.
“Using AI can help with the identification of new therapeutic targets, novel molecule design and the analysis of clinical trials. Nowadays, these methods are also applied in laboratory software tools to extract the most relevant information and enable scientists to make faster decisions as the data is being generated."
"Applying these AI tools will reduce the time needed to find and extract relevant information during R&D and beyond, thereby increasing our employee productivity and helping us to develop better drugs faster.”
AI can optimise the processing of a range of data types
There’s a significant amount of labour involved in data handling and, often, cross-referencing is difficult and time-consuming to do manually.
However, AI offers a solution to this, allowing the processing of many data forms simultaneously, helping to find connections where humans may miss them. “Developing specific AI tools will optimise our access to high quality data, including data governance, quality processes and digital identity management, amongst others,” emphasises Mr Fernández.
Enhancing the role of AI in pharma
Although AI’s influence has grown significantly, Almirall feels it will only continue to expand into the pharmaceutical landscape. Francesc elaborates: “We believe AI can be further utilised in the drug development process as more and more high-quality data is constantly being released and better AI methods will be developed."
"Considering the current state and pace at which AI technologies are progressing, it’s quite sensible to believe that many of these activities will be performed by these tools very soon … and they will act as advisers to scientists and other staff.”
