Research
Our multidisciplinary research team integrates a variety of approaches
Our multidisciplinary research team integrates a variety of approaches (i.e., neurophysiological, behavioral, biochemical, engineering, computational) to analyze and identify patterns of electrical activity in brain networks in awake, freely behaving mouse models under normal and disease conditions.
Neurophysiology/electrophysiology describes the electrical recording of neural activity. We aim to develop circuit interventions that may be translated for novel treatment of psychiatric disorders in humans.
Data Science & Bioinformatics
In collaboration with the Carlson Lab, we have an ongoing effort to advance machine learning and artificial intelligence (AI) methods to analyze the brain. Specifically, we use complex computational algorithms to identify relevant network models of the brain from our data, and we use these learned models to help us answer scientific questions, and help us design future experiments. Achieving this goal requires the development of novel algorithms and mathematical approaches in addition to close collaboration with neuroscience experts. These techniques push the state-of-the-art in the fields of machine learning and network neuroscience.
Neuroscience & Neuroengineering
Utilizing in vivo multi-site electrophysiology and other tools to monitor and modulate brain function. In conjunction with machine learning, we record neural activity in behaving mice and study how brain regions interact with each other across space and time within brain circuits that we believe underlie various emotional and cognitive states.
-
Depression is a heterogeneous psychiatric disease, meaning there are many different ways that a person with depression may look, feel, or behave. Therefore, identification of brain network features that are distinct and conserved across different models of depression could allow us to target predominant features with a “cure-all” or tailor treatments to specific forms of depression. To identify conserved and distinct features, we use several different mouse models of depression, including genetic, sleep disruption, hormonal, and various forms of chronic stress. We then extract and compare learned networks between the different models to pinpoint similarities and differences. These models could provide the foundation for better understanding the heterogeneity of depression in humans and developing translational treatments.
-
We adopt a hypothesis- and data-driven approach to characterizing the spatiotemporal dynamics underlying negative valence states (anxiety and fear). We combine multi-site electrophysiological recordings along with machine learning and pattern recognition techniques to identify neural signatures relevant to these emotional states. These signatures are then examined in various contexts, including but not limited to: activation following acute/chronic stress, utilization in disease states (bipolar mania, depression, substance abuse), and differences across sex/genotype.
-
We are interested in learning the brain networks that guide motivated behaviors as well as how disruptions of network activity may be related to reward-related disease states, including drug addiction, depression, and bipolar disorder. Using tools that measure and manipulate brain activity in mice and machine learning approaches that identify related patterns of activity across numerous brain regions, we aim to understand network-level mechanisms underlying reward-related behaviors and to identify new targets for therapeutic intervention.
-
We leverage mouse models of neuropsychiatric conditions alongside multi-site electrophysiology to identify brain networks underlying social behaviors using machine learning and data science tools. Closed-loop optogenetics is a technique that enables optogenetic stimulation for biochemical neuromodulation, provides cell-type specificity, and is time-resolved to the millisecond timescale. LinCx (pronounced “links”) is an ongoing project which aims to develop a biochemical electrode which would eliminate the need for microwire implantation. We utilize these brain networks to dissect network mechanisms using neural manipulations (e.g., closed loop optogenetics and LinCx).
Understanding these mechanisms will enable the development targeted treatments to restore or enhance functioning and promote mental well-being.
-
We aim to link electrical measures of brain activity and connectivity identified in the above contexts with biological measures of brain function– including activity of different cell types and levels of neurotransmitters and neurochemicals. By simultaneously integrating in vivo multi-site electrophysiology in freely-behaving mice with approaches such as calcium imaging and fiber photometry, we study how brain network activity relates to these biological phenomena, as well as behavioral changes both under healthy conditions and with pharmacological manipulations.
Biochemistry & Protein Engineering
We endeavor to generate, characterize, and optimize engineered gap junctions for use as novel neuromodulatory tools for mammalian circuit manipulation – both as standalone devices and as integral parts of multi-component systems. We employ numerous techniques pertinent to molecular biology, biochemistry, cell biology, and computational biology to achieve these goals. Ultimately, our work aims to contribute novel and transformative tools to advance the study and treatment of psychiatric illnesses.
