The Children's Tumor Foundation (CTF) has been selected for Anthropic's AI for Science rare-disease research program. Through the program, CTF will receive Claude credits to develop an AI-assisted pipeline that uses gene-expression data and scientific literature to identify potential drug targets.
CTF will build and validate the pipeline using publicly available data related to neurofibromatosis type 1 (NF1), NF2-related schwannomatosis (NF2-SWN), and schwannomatosis (SWN). The team plans to use the resulting approach as a foundation that can eventually be adapted for other rare diseases facing similar data limitations.
Turning Existing Knowledge Into Actionable Evidence
For rare diseases, the challenge is not always an absence of scientific knowledge. Valuable data and evidence already exist, but they are often scattered across small datasets, disease-specific studies, and a vast body of scientific literature. Unless that information can be connected and interpreted, it is difficult to translate it into decisions about where to focus drug-discovery efforts.
CTF's project will address this challenge by building a documented, reusable pipeline that can identify regulatory genes that may offer opportunities for therapeutic intervention.
The pipeline will draw on multiple scientific databases to help researchers evaluate potential drug targets: exploring whether existing drugs might work, whether the evidence holds up in related rare diseases, whether there are safety concerns, and whether the findings are well-supported enough to act on.
Building AI Into the Drug-Discovery Process
This project advances CTF's ongoing work to apply artificial intelligence across drug discovery and development.
AI can help researchers analyze large volumes of scientific data, identify connections that may be difficult to detect manually, and evaluate potential opportunities more efficiently. But its value depends on the quality of the underlying data and the rigor of the process around it. The CTF pipeline is being designed not only to identify promising targets, but also to recognize and document when the evidence cannot support a reliable conclusion.
The award is part of Anthropic's broader AI for Science program, which provides researchers with access to Claude resources like tokens and Claude Science to explore how AI can accelerate scientific discovery.
By building and testing this pipeline with NF data, CTF aims to turn existing scientific information into clearer, evidence-backed opportunities for drug discovery, helping researchers identify where to focus next and supporting a faster, more informed path toward treatments.