A groundbreaking study has unveiled an AI model capable of predicting the degeneration of neural networks in Amyotrophic Lateral Sclerosis (ALS), offering a new perspective on this devastating disease. This innovative research, a collaborative effort between the University of St Andrews, the University of Copenhagen, and Drexel University, could revolutionize how we understand and combat ALS. But, what exactly does this mean for patients and the future of ALS research? Let's dive in.
This study, published in Neurobiology of Disease, introduces computational modeling as a powerful tool to complement existing methods, such as animal and in vitro studies.
Motor Neuron Disease (MND) encompasses a group of illnesses affecting motor neurons in the brain and spinal cord. ALS, the most prevalent subtype, is often referred to as Lou Gehrig's disease or Maladie de Charcot in some countries. ALS affects roughly 2 out of 100,000 individuals annually worldwide, translating to approximately 200 new diagnoses each year in Scotland alone.
Typically, ALS begins in the spinal cord, impacting motor neurons and specific neural circuits. Early symptoms often manifest as muscle weakness, stiffness, and cramps.
Traditionally, scientists have relied on animal models, like genetically modified mice, to study ALS. Researchers observe how the disease progresses in these models, focusing on specific time points due to resource constraints.
However, the beauty of computational models lies in their ability to predict what happens between these time points, providing a more comprehensive understanding of disease progression. Furthermore, these models allow researchers to conduct controlled experiments, modifying a single variable at a time, unlike animal models, where multiple factors are always at play.
And this is the part most people miss... Computational models can also predict how neural circuits might respond to treatments, guiding future preclinical studies in mice.
The researchers in this study utilized biologically plausible neural networks. Unlike the everyday neural networks used for facial recognition or answering questions with ChatGPT, these networks communicate using spike signals, mimicking the nerve cells in our nervous system. These models are structured based on known biological data about the cells and connections in the spinal cord.
The models, developed by researchers from the School of Psychology and Neuroscience, use mathematical equations to calculate the excitability of each neuron. When a neuron receives a spike (an electrical impulse), it changes its excitability, and if it's excited enough, it will spike, passing information to the next neuron. These neurons are grouped into populations and connected based on biological data to construct the network.
Co-author Beck Strohmer from the University of Copenhagen explained, "During ALS, neurons die, and communication breaks down. We model this by removing neurons and reducing connections, allowing us to model disease progression. We can also model and test treatment strategies."
Dr. Ilary Alodi, a Reader at St Andrews School of Psychology and Neuroscience, added, "Hypotheses generated by models need to be tested on animal models because it is impossible to model all the complexities of a biological system. In this study, we predicted that the applied treatment strategy in the model would save a specific population of neurons. We then looked at this neuron population in the treated mice and found that hypothesis held true."
These results highlight that while caution is necessary, models can effectively guide experimental research.
This means animal experimentation can be refined, focusing research efforts on specific areas and time points.
Dr. Alodi also noted, "We are now starting to apply these models to specific brain areas to understand how neuronal communication changes during dementia, which is an exciting new research direction for our lab."
Controversy alert: While these AI models show promise, some may argue about the ethical implications of using AI in healthcare. What do you think? Are you optimistic about the potential of AI in ALS research, or do you have concerns? Share your thoughts in the comments below!
