Insilico Medicine Identifies Alzheimer’s Targets Using AI

Hong Kong, September 1, 2026

AI Analysis Maps Molecular Changes in Neurodegeneration

The research team investigated genes associated with mitochondrial function and NAD⁺ metabolism across 12 healthy brain regions, as well as in cerebrospinal fluid and blood. Researchers compared molecular patterns associated with healthy aging with changes observed in four major neurodegenerative diseases: Alzheimer’s disease, Parkinson’s disease, Huntington’s disease and amyotrophic lateral sclerosis (ALS). The analysis showed that widespread alterations in pathways linked to mitochondrial function and NAD⁺ metabolism can emerge during the early stages of neurodegenerative disease. Importantly, several of these molecular changes associated with Alzheimer’s and Parkinson’s disease were also detectable in blood samples, indicating potential applications as early blood-based biomarkers. The findings are significant because neurodegenerative diseases remain challenging areas for drug discovery, partly because of their complex biological mechanisms and the difficulty of translating molecular discoveries into effective therapies. By integrating large-scale human data with computational analysis, the researchers sought to identify biological pathways that could provide actionable opportunities for future research. The study supports the use of multi-modal biological data and AI as tools for prioritizing disease mechanisms and potential therapeutic targets earlier in the drug-development process.

Insilico Medicine Prioritizes Five Alzheimer’s Targets

To investigate potential therapeutic opportunities, Insilico Medicine applied its proprietary PandaOmics AI biology platform to more than 100 candidate genes associated with NAD⁺ and mitophagy pathways. The platform prioritized five potential Alzheimer’s disease targets: ULK1, OPA1, LAMP2, MFN1 and ATP6V0E1. These targets were selected for further investigation based on the computational analysis and their potential relationships with mitochondrial quality control and neurodegenerative pathology. The researchers subsequently moved beyond computational predictions and conducted preclinical experimental validation using multiple biological systems. The prioritized targets were evaluated in C. elegans, a human Tau-mutant cell line and APOE4/4 iPSC-derived cortical neurons. This experimental approach provided evidence that functional modulation of the identified genes could influence disease-associated biological processes. In particular, activation of OPA1, a gene involved in mitochondrial fusion, increased cell viability and significantly reduced Tau phosphorylation in APOE4/4 cortical neurons.

Study Links Mitochondrial Quality Control to Alzheimer’s

Additional experiments strengthened the connection between mitophagy and neurodegeneration. Researchers found that reducing the activity of key mitophagy drivers, including MFN1 and LAMP2, worsened Tau aggregation. These observations suggest that maintaining mitochondrial quality-control mechanisms may play an important role in limiting pathological processes associated with neurodegenerative disease. The results provide a biological foundation for investigating mitochondrial pathways as potential therapeutic intervention points in Alzheimer’s disease and related disorders. The study also illustrates an emerging approach in AI-enabled drug discovery, in which computational predictions are paired with laboratory experiments rather than being treated as standalone evidence. Insilico Medicine said the collaboration demonstrates how AI-based biological analysis can help identify potential targets while experimental systems provide a mechanism for testing their relevance. The researchers’ findings could support future development of Alzheimer’s disease biomarkers and therapeutic strategies, although the identified targets remain at the research and preclinical stage and require further investigation before any clinical application can be established.

Source: Insilico Medicine press release