Summary: This blog explores how AI and machine learning can accelerate neurological disease research by leveraging large-cohort study data to uncover disease mechanisms, identify novel drug targets, and drive data-driven therapeutic advancements.
Neurological disorders, such as Alzheimer’s and Parkinson’s diseases, multiple sclerosis, dystonia, spasticity, sialorrhea, and other conditions, represent a significant healthcare burden globally. In particular, neurodegenerative diseases like Alzheimer’s disease, dementia and Parkinson’s disease are expected to increase in prevalence in the upcoming decades due to the aging of the global population. These conditions are notoriously difficult to diagnose and treat due to a lack of understanding of disease mechanisms, early-stage diagnostic tools and confirmed disease targets. Neurodegenerative processes often take place over decades, suggesting a complex interplay of factors that lead to disease development and progression, which further complicates the task of neurological target identification.
These challenges contribute to the current lack of effective treatment options for neurological conditions. For example, existing treatments for Parkinson’s disease, such as levodopa and amantadine, are more effective in the early stages of the disease, with effectiveness diminishing over time. While recent successes such as the approval of AbbVie’s VIALEVO™ for adults living with advanced Parkinson’s disease have shown promise, much remains to be done. For example, there are a lot of open-ended questions in identifying effective targets for neurodegenerative disease treatments, such as:
1) What triggers the onset of the disease?
2) When do these changes begin?
3) How can we diagnose it early and begin timely intervention?
4) Are neurological changes reversible and can we treat them?
5) What are the potential therapeutic targets?
At Ardigen, we are actively tackling these and other research questions. Let’s take a look at how Artificial Intelligence (AI) and Machine Learning (ML) approaches are changing the landscape of neurological disease research.
Artificial intelligence takes a shot at neurodegenerative diseases
As with other disease areas, AI has infused new energy into the development of neurotherapeutics. A recent announcement that insitro received $25 million from Bristol Myers Squibb for identification and validation of the first novel target for amyotrophic lateral sclerosis (ALS), underscores the utility of AI and ML approaches for neurology research. With the capacity to mine and analyze vast datasets from genomics, proteomics and patient records, these approaches can improve the understanding of disease mechanisms, discover relevant biomarkers and identify novel drug targets.
Several large-scale research initiatives to collect neurological disease data have been undertaken. These data repositories contain a wealth of biological information that can be leveraged for data mining using ML and AL. For example, the Parkinson’s Progression Markers Initiative (PPMI) is a large-scale research project launched by the Michael J. Fox Foundation (MJFF) to identify biomarkers of Parkinson’s disease (PD) progression. PPMI collects comprehensive data across biological, clinical, imaging, genetic and digital domains, which can be analyzed using machine learning methods to identify patterns or biomarkers for predictive modeling of PD progression.
Another important resource is Fox Insights, an online clinical study sponsored by the MJFF in partnership with 23andMe that gathers data directly from people with Parkinson’s disease (PD) and healthy volunteers. This is a rich collection of real-world clinical data where participants share their personal experiences, health information, and symptoms via online surveys, creating a large, real-world dataset. This data can be used in the development of predictive models to improve clinical trials and improve treatments for individuals with PD.
The ASAP Initiative is a global research program that fosters collaborative, open-science research on PD. It focuses on sharing data and biological insights with researchers across the world One of the key projects sponsored by the ASAP Initiative is the GP2 – Global Parkinson’s Genetics Program, aimed at identifying genetic risk factors for PD by collecting and analyzing genetic data from diverse populations around the world. It contains genetic data from over 150,000 individuals, including people with Parkinson’s disease and healthy controls, which could be used to derive associations and identify genetic biomarkers of PD to drive personalized treatment approaches.
These initiatives offer invaluable orthogonal omics data sources and patient data for exploration using advanced computational tools to establish proof-of-association between specific biological factors and PD outcomes. Collaboration with these organizations could uncover hidden patterns in biological data, develop predictive models for disease progression and identify potential drug targets. As we set our target on developing new innovative approaches to tackling the treatment of neurodegenerative disease, AI and ML tools will become indispensable in analyzing the vast array of collected data.
Case study: Validating and expanding a set of potential targets
Ardigen has recently supported a company in a neurodegenerative disease case. Using experimental and clinical Parkinson’s disease data from the Michael J. Fox Foundation database, we analyzed multiple lines of evidence from orthogonal omics data sources and patient data to establish a robust proof of association between specific biological factors and disease outcomes. Additionally, we used the data to improve the understanding of the underlying mechanisms of action for these associations by employing multiple computational methodologies.
By combining the Michael J. Fox Foundation data with additional curated data from the public domain, Ardigen was able to provide insights into promising mechanisms of the disease connected to lipid metabolism. Additionally, we were able to validate known targets, expand the list of potential candidates, and stratify the network of associated genes based on their impact, significance, and influence on Parkinson’s disease.
As a result, these findings enabled the development of a novel research strategy based on insights into the metabolic links to Parkinson’s disease, providing an expanded list of potential targets.
Tackling neurology research questions through advanced computational approaches
AI and ML methods provide new ways to address critical questions in neurology research. By leveraging data from large-scale repositories like those from the Michael J. Fox Foundation, Ardigen applies advanced analytical tools to explore disease mechanisms, identify potential biomarkers and validate therapeutic targets.
Collaborating with organizations collecting extensive biological and clinical data allows for a deeper understanding of neurodegenerative diseases and the development of innovative strategies for early diagnosis and treatment. Reach out to one of Ardigen’s experts to discover how your program can benefit from utilizing large-scale neurological disease data.
