Artificial intelligence can revolutionize drug discovery. To harness its full potential, pharmaceutical companies need effective big data strategies.
Strategic Thinking
In today’s highly digitized workplace, the generation and aggregation of huge amounts of data has become the norm. Organizations of all sizes are increasingly harnessing the immense potential of “big data” to gain valuable insights, make informed decisions and drive research and innovation.
The pharmaceutical industry, which is continually searching for groundbreaking ways to improve the drug discovery process, is one such industry. Novel drug discovery is an intricate, time-consuming and resource-intensive process that involves numerous stages—from target identification through to clinical trials.
Processing big data using AI-empowered systems has the potential to transform every step of this process. But drug companies must have an effective big data strategy in place to leverage the full benefits. Nicola Richmond, Vice President of AI at London-headquartered Benevolent AI, a clinical-stage AI-enabled drug discovery and development company, explained:
“AI methods, which are supremely data-hungry, are 100% reliant on high-quality data that is relevant to the task at hand.
Big pharma possesses gold mines of amazing data, but it is often siloed due to its size, age and complexity. If pharmaceutical companies want to advance drug discovery using AI, they absolutely must have a joined-up data strategy in place that supports integration and analysis.”
Improved Healthcare Outcomes
Leveraging AI approaches on sufficient quantities of information-rich, joined-up multimodal data can help drug companies find the best medicines to improve healthcare outcomes.
A solid data foundation of high-quality, integrated biomedical data can be mined for new insights, such as repurposing known drugs for new indications, or for building and using biospecific AI models to identify potential novel therapeutic targets and to discover efficacious drug candidates that have a higher chance of success in clinical development.
Nicola Richmond, who recently gave a presentation on data strategy at the Lab of the Future 2023 Congress in Amsterdam, added:
“In January 2020, as the COVID-19 pandemic spread, BenevolentAI set out to find an existing drug that could be repurposed as a COVID-19 treatment.
Although we designed our technology to develop new drugs for disease—not identify new uses for existing medications—we could make this pivot thanks to our comprehensive data foundations and flexible, disease-agnostic approach.
The result was the repurposing of Eli Lilly’s baricitinib, which reduced deaths in hospitalized COVID-19 patients by 38%.”
BenevolentAI has invested heavily in multimodal, biomedical data ingestion, harmonization and integration. The company’s data foundations comprise over 85 different sources, spanning a range of modalities, including patient-level data (such as electronic health records, genetics, and “omics”), peer-reviewed biomedical literature, patent data, compound-level data and clinical trials data.
Huge Potential For Precision Medicine
Drug development software driven by AI can revolutionize many different pharmaceutical discovery processes. Leveraging the power of big data, it can even underpin a transformation towards “AI-first” drug discovery and development, reducing the need for physical experiments.
There is huge potential, for example, in the field of precision medicine, which aims to provide targeted treatments tailored to an individual patient. Big data can play a critical role in achieving this goal by providing vast amounts of information about patient populations and their unique characteristics, such as genetic makeup, lifestyle factors and medical history.
By analyzing large-scale data sets, researchers can develop algorithms that predict how a patient will respond to a particular treatment, which can help to avoid exposing them to drugs that are unlikely to be effective. Richmond said:
“We’re now seeing powerful, multimodal AI models emerge onto the scene that understand text and images.
I see a future world where humans and multimodal AI work in synergy to understand a patient’s disease, recommend strategies for reversing disease pathology and help design medicines with the desired properties from the outset.
But we’ll only get there with high-quality, multimodal, relevant data generated and collected as a result of cutting-edge data strategies.”
