Our group collaborates with neuroscientists to explore the functioning of neurons and networks, using both invertebrate and vertebrate model systems. Based at the University of Missouri (MU), our neuroscience collaborators include faculty from MU and other institutions.
The Lab focus is on neural engineering research at multiple levels, in close collaboration with neuroscientists. Some of our computo-experimental collaborations are listed below:
- cellular/circult level: neural correlates of fear and extinction memories
cellular/circuit level: neural oscillations, synchrony and information processing
celullar level: neuroadaptations in the PFC-NAc perisynapse due to cocaine
intra-cellular level: LTP/LTD in the accumbal PSD due to cocaine
behavioral/clinical levels: Machine learning, AI models, and Big-Data in neuro-psychiatry
all levels: analytical insights into structure-function relationships, including via reduced order models
all levels: software automation, including cyberinfrastructure tools for high performance computing, big data… in neuroscience
all levels: development of a regional neuro-community via the portal CyNeuro.org
Our research is an interdisciplinary collaboration between the following labs and research groups:
- Nair Lab – MU College of Engineering
- Kalivas Lab – Medical University of South Carolina
- Pare Lab – Rutgers University
- Quirk Lab – University of Puerto Rico School of Medicine
- Schulz Lab – MU Biological Sciences
Systems and Computational Neuroscience
The Nair lab projects involve reverse engineering the brain circuits in invertebrates and vertebrates, at intracellular, cellular and systems levels, in close association with neuroscientists and biologists. We model a neuron as a nonlinear electrical circuit and combine many neurons to form networks. Using biologically realistic and reduced order models, we use system theoretic concepts to investigate how such neurons/network circuits implement functions in normal and pathological brains. Such functions include neurocomputational properties, adaptation, learning (LTP/LTD, etc.) and robustness….for these nonlinear dynamic circuits. We also study genesis and control of neural oscillations using multiple model types, and work with both neuroscientists and clinicians to optimize opto- and electrical- brain stimulation algorithms.
Some of our present application areas include anxiety disorders – how are conditioning and extinction fear memories acquired and stored in the amygdala and the associated cortical structures? How does context modulate fear and extinction?, What is the role of theta and gamma oscillations and how can we design optimal brain stimulation algorithms to alleviate anxiety disorders? …. neuroplasticity due to cocaine: what are the neuroplasticity mechanisms that might explain known cellular adaptations due to cocaine in the PFC-NAc glutamatergic pathway? What are possible mechanisms of LTP/LTD in the accumbal PSD? …. machine learning models in neuroscience: e.g., detecting prescription opioid abuse using a large hospital dataset
Biologically realistic modeling of neural circuits will provide a fundamental understanding of the underlying brain mechanisms in both health and disease (e.g., PTSD, anxiety disorders, neuroplasticity due to cocaine). Such models will also help lay the groundwork for innovative pharmacological, psychotherapeutic and other treatments by permitting rapid in-computer experimentation.