St. Jude Hospital Launches Combocat to Revolutionize Drug Discovery

Researchers at St. Jude Children’s Research Hospital unveiled a groundbreaking screening platform named Combocat, designed to streamline the discovery of effective drug combinations. This innovative approach addresses the increasing complexity of drug discovery, particularly for diseases like cancer that often require multi-drug regimens for effective treatment. The findings were published in the journal Nature Communications.

As the number of available drugs and potential combinations has surged, traditional screening methods have become impractical. Paul Geeleher, Ph.D., a senior co-corresponding author from St. Jude’s Department of Computational Biology, emphasized the need for an efficient solution. “The field of drug discovery has lacked a way to deal with the sheer number of potential combinations,” he stated. “We designed Combocat to use minimal resources and enable scientists to test massive numbers of drug combinations, rapidly nominating those most likely to have synergistic effects to explore further.”

Combocat employs a combination of machine learning and specialized liquid handling technology to facilitate large-scale combination screening. The platform’s capabilities were showcased through testing 9,045 pairs of drugs against a neuroblastoma cancer cell line, revealing multiple drug pairs with strong synergistic effects. The top findings were confirmed through additional experiments, demonstrating Combocat’s potential to efficiently identify promising drug combinations.

Combining Technology and Innovation

To achieve its ambitious goals, Combocat integrates miniaturized drug dispensing with advanced machine learning techniques. Sonic technology plays a crucial role in drug dispensing, allowing for precise and efficient experimental layouts. Charlie Wright, Ph.D., the first co-author, noted, “We incorporated acoustic liquid handlers that use sound waves to transfer tiny droplets of drugs very precisely. They use the exact minimum of each liquid you need, allowing for the use of far less material than conventional pin or pipette-based techniques, and increasing the number of testable combinations.”

The platform operates in two distinct modes: “dense mode” and “sparse mode.” In dense mode, researchers measure every possible dose pairing for each drug combination. In contrast, sparse mode utilizes machine learning to predict results from a smaller subset of data, enhancing resource management. The sparse mode model was trained using data from hundreds of drug combinations tested in dense mode, yielding highly consistent predictions when compared to actual measurements.

“We created two screening ‘modes,’ with something of a tradeoff between them,” explained Geeleher. “We optimized the sparse mode approach for scale, but it trades detail for efficiency, while we optimized the dense mode to obtain ultra-reliable measurements, which can’t scale. However, we showed that Combocat can combine them to analyze more combinations and validate them more rapidly than traditional approaches.”

A Legacy of Innovation in Drug Discovery

Combocat continues a legacy of innovation in drug combination therapy at St. Jude, providing a powerful tool for researchers not only in oncology but across various diseases. Geeleher expressed optimism about the platform’s potential impact, stating, “We’ve created a platform that’s free, open-source, and highly usable that could become a strong standard in the drug combination discovery field.”

The implications of Combocat are significant, as it can help expedite the identification of safe and effective drug combinations, potentially leading to new treatment regimens that could change clinical practices. The platform is accessible to researchers worldwide, fostering collaboration and advancing the field of drug discovery.

For more information on Combocat, visit [Combocat’s website](https://www.stjude.org).

William C. Wright et al., “An open-source screening platform accelerates discovery of drug combinations,” Nature Communications (2025). DOI: 10.1038/s41467-025-66223-8.