AI in Drug Discovery: Quality Data and Smart Questions Are Key

The promise of artificial intelligence (AI) in drug discovery is immense. It could revolutionize how we develop new treatments, making them faster, more effective, and accessible to millions. However, as experts at the 2026 STAT Breakthrough Summit East emphasized, the success of AI in this field hinges on two critical factors: high-quality data and well-posed research questions.

At the summit, Gonçalo Abecasis, senior vice president and chief genomics and data science officer at Regeneron Genetics Center, joined reporter-at-large Damian Garde to discuss these challenges. The conversation revealed that while AI holds tremendous potential, its effectiveness is directly tied to the quality and integrity of the data it processes.

The Role of Data in AI-Driven Drug Discovery

Abecasis began by sharing his journey into genetics and technology. As a teenager with a burgeoning software business, he initially thought he knew everything about computer science. But when it came time for college, he decided to explore something entirely new-genetics. This decision led him to Oxford, where he developed technologies for genetic analysis in the lab.

The tools available at the time were inefficient and cumbersome, which inspired Abecasis to improve them. “I could see that there was a lot of room for innovation,” he said. “And that’s how I ended up merging my interests in technology and biology.”

In drug discovery, data quality is paramount. Abecasis explained that AI models are only as good as the data they are trained on. Poor-quality or incomplete data can lead to flawed insights and ineffective drug candidates. On the other hand, high-quality, comprehensive data sets can significantly enhance the accuracy and reliability of AI-driven discoveries.

“Data integrity and scale are crucial,” Abecasis emphasized. “We need large, diverse data sets that capture a wide range of genetic variations and health outcomes. This helps us identify patterns and correlations that might not be visible with smaller or less diverse data sets.”

What Comes Next

The future of AI in drug discovery is promising but requires ongoing effort to address current challenges. One key area is improving data collection methods. Abecasis highlighted the importance of standardized protocols for collecting, storing, and analyzing genetic and health data. This standardization ensures that data from different sources can be reliably integrated and analyzed.

Another critical aspect is posing the right research questions. “It’s not just about having more data; it’s about asking the right questions,” Abecasis said. “We need to frame our hypotheses and research objectives in a way that leverages the strengths of AI.”

For example, instead of simply looking for genetic markers associated with a disease, researchers can use AI to explore complex interactions between genes, environmental factors, and lifestyle choices. This holistic approach can lead to more nuanced understanding and better-targeted treatments.

The collaboration between technology experts and biologists is also essential. Abecasis stressed the importance of interdisciplinary teams that combine expertise in data science, genomics, and clinical research. “These collaborations are crucial for translating AI insights into real-world applications,” he said.

As AI continues to evolve, its impact on drug discovery will grow. By focusing on high-quality data and well-posed questions, researchers can unlock new possibilities for treating diseases and improving public health. The future of medicine may well be shaped by the synergy between human ingenuity and machine intelligence.