AI-driven discovery of in vivo chemical probes for understudied targets for the treatment of alzheimer's disease and related disorders (ADRD) - alzheimer's disease and related dementias (ADRD) are…

The National Institute on Aging awarded Fox Chase Chemical Diversity Center Inc. $1,074,362 on March 1, 2026, under the Aging Research program (CFDA 93.866) to discover in vivo chemical probes for understudied Alzheimer's disease targets using artificial intelligence and machine learning.

Fox Chase will apply AI and machine learning to identify novel chemical probes targeting proteins relevant to Alzheimer's disease and related dementias. The work encompasses five key aims: target selection aligned with the TREAT-AD consortium, virtual screening of Fox Chase's compound libraries to identify novel compounds with desired binding activity, assay development and testing of lead compounds and structural analogs, AI-driven algorithm development to optimize selective leads, and medicinal chemistry refinement with biochemical and functional characterization including ADME assessment. Fox Chase will evaluate approximately 15 proteins annually from the 900 proteins listed on the TREAT-AD website and advance five targets through structure-activity relationship (SAR) development activities annually. Chemical probes developed will be made publicly available through TREAT-AD and shared with the scientific community.

Performance takes place in Doylestown, Pennsylvania. The period of performance extends from March 1, 2026, through November 30, 2030. The National Institute on Aging's Aging Research program supports biomedical and behavioral research directed toward understanding aging and diseases including Alzheimer's disease and related dementias.

AI-driven discovery of in vivo chemical probes for understudied targets for the treatment of alzheimer's disease and related disorders (ADRD) - alzheimer's disease and related dementias (ADRD) are debilitating diseases that are increasing in prevalence being the most common form of dementia worldwide. Current therapies are primarily palliative in nature except for the recent amyloid antibodies which show variable and incremental efficacy in slowing disease progression. We here apply the emerging tools of artificial intelligence (AI) and machine learning (ML) to aid in the discovery of in vivo chemical probes for understudied targets in ad research. The project is closely aligned with the NIA- funded ad initiatives accelerating medicines partnership in alzheimer's disease target discovery and preclinical validation project and the emory-sage-SGC-JAX target enablement to accelerate therapy development for ad (TREAT-ad) center. We will examine all 900 proteins currently listed on the TREAT-ad website, evaluate 15 per year using virtual screening from our compound libraries, and advance five of these through a series of SAR development activities leading to in vivo chemical probes which are made publicly available through TREAT-ad and shared with the scientific community. We will leverage validated virtual screening workflows for novel target discovery, a process for which we have extensive experience. Aim 1 is target selection closely aligned with TREAT-ad. Aim 2 is virtual screening for novel compounds with desired target binding activity. Aim 3 is assay development and testing of top hits from aim 2 plus structural analogs. An ad hoc target project team will be created for each target that emerges from aim 2. Aim 4 incorporates novel AI-driven algorithm development and training for optimizing selective leads from aim 3 hits. Aim 5 is medicinal chemistry refinement of the AI/ML outputs and biochemical and functional characterization of chmical hits including adme characterization suitable for the identification of probes for in vivo studies. A HIT review steering committee (HRSC) will review the hits and leads that are discovered during the course of this project to provide guidance as to design and interpretation of studies as well as which hits and chemotypes to advance further. Finally, aim 6 is public dissemination of the structures and activities of new in vivo chemical probes for understudied targets in ad research. Small molecule chemical probes are validated for further proof of concept research if they have binding or functional activity at targets of interest to the extent of ≤100 NM (PREFERRED) and in vivo PK in mice with a T1/2 of ≥60 mins (PREFERRED), %F (oral bioavailability) of ≥20% and a brain:plasma ratio of ≥0.5. In summary, 15 targets per year will go into a full virtual screen. Based on the data, five targets per year will go to experimental validation and SAR development. All of the data will be released on the TREAT-ad website so others can pursue the 10 targets per year that were not followed up on. We will use modern methods in computational chemistry including AI and ML to discover and characterize new in vivo chemical probes for understudied targets in ad research. IT is anticipated that these in vivo chemical probes will be widely used by researchers in the ad scientific community to understand the function of promising new targets.