Sarah Ateaque, a scientific specialist at Causaly, delivered a presentation on enhancing drug discovery using human-centric generative AI. She began by outlining the significant challenges in drug development, noting that it typically took 10 to 15 years and cost $1-2 billion to develop a new drug. She highlighted that less than one in 10,000 potential drug compounds successfully made it through clinical trials, with 52% of drugs failing due to lack of efficacy, often attributed to poor target selection during the pre-discovery phase.
Ateaque then introduced Causaly, an AI solution designed to support and accelerate biomedical discoveries. She explained that traditional literature search methods using tools like PubMed were outdated and inefficient, leading to data overload, bias, loss of hidden evidence, and poor traceability. Causaly, on the other hand, integrates multiple data sources to create a high-precision knowledge graph, allowing researchers to receive direct answers to their questions and view all related evidence quickly.
To illustrate Causaly's capabilities, Ateaque presented a case study involving a discovery scientist named Jo, who aimed to identify promising drug targets for Huntington's disease. Using Causaly, Jo identified 1,300 potential targets and prioritised them based on genetic mutations and novelty. Causaly provided an interactive dendrogram format for exploring targets, generated AI summaries with linked citations, and allowed for comparative analysis of targets.
Ateaque demonstrated how Jo used Causaly to identify and validate two candidate targets: the emerging noncoding RNA microRNA 615 and the human mutated JPH3. Causaly's multi-hop functionality helped Jo discover hidden links between these targets and Huntington's disease, further strengthening their validation. Jo also used Causaly to understand key assays for validating the targets in the lab and shared her findings with her team through Causaly workspaces.
In conclusion, Ateaque emphasised Causaly’s ability to enable scientists to complete their tasks more efficiently, saving time and reducing bias. She quoted, "Scientists will not be replaced by AI, but they may be replaced by scientists that use AI," highlighting the importance of integrating AI into scientific research.
