top of page

Research Overview and Philosophy

We are an interdisciplinary and curiosity-driven group interested several interrelated questions in basic neuroscience. We are unapologetically interested in fundamental questions guided by the steadfast belief that progress in basic science is the foundation for practical applications. We combine computational, theoretical and experimental approaches in this quest. Current efforts are focused on two major themes. The first theme involves using single-cell genomics to gain a deeper understanding the molecular basis of the diversity of neurons across the brain, understanding the development and evolution of neural diversity, and its implication for neurodegenerative disease. The second theme uses continuum physics to understand ion transport and electrostatic effects near membrane interfaces, and exploring its implications for neuronal signaling.

Our work is highly collaborative, and we have active collaborations with experimental neuroscientists, molecular biologists and clinicians. Our research has been supported by several federal and private foundations including the National Institutes of Health, National Science Foundation, Glaucoma Research Foundation, Chan Zuckerberg Initiative, McKnight Foundation, and BrightFocus Foundation.  

Below, we highlight a few ongoing projects and previous work.

Understanding the genomic basis of neuronal diversity

The brain is comprised of diverse types of neurons that vary in form and function. We have pioneered approaches in single-cell genomics to chart the molecular diversity of cell types in the retina, the thin neural film in the eye that initiates vision. In collaboration with neuroscientists, we have also shown that these molecular atlases can be harmonized with morphology and physiology. Our single-cell genomic analyses of the mouse retina can be found in multiple papers over the past several years. For example, Shekhar et al., Cell, 2016 and Tran*, Shekhar* et al., Neuron, 2019 describe comprehensive atlases of retinal bipolar cells and ganglion cells in mice, respectively. Goetz et al., Cell Rep, 2022 describes collaborative work to unify morphology, function and transcriptomics for retinal ganglion cells in mice.  Peng*, Shekhar* et al., Cell, 2019 and Yan et al., Sci. Rep., 2020 describe the classification of the primate retina in macaques and humans. A detailed review of this work is presented in Shekhar and Sanes, Ann. Rev. Vis. Sci., 2021.

image_edited.jpg
image.png

What is the role of nature and nurture in the development of cell types?

A central question in developmental neuroscience is how the diverse types of neurons in the brain are generated, and how they assemble into highly elaborate circuits that govern our sensory perceptions, information processing and behavior. We have attempted to address this question in two different parts of the visual system, the retina and the cortex.

In the retina, we analyzed the developmental diversification of retinal ganglion cells (RGCs), the neurons that connect the eye to the brain. Using a computational approach rooted in Optimal Transport theory, we have shown that transcriptomically distinct RGC types arise gradually in mice over a protracted period (Shekhar et al., eLife, 2022). In contrast, we have shown that this process happens rather rapidly in zebrafish, and leads to the emergence of types that regulate highly specific visual behaviors (Kolsch et al., Neuron, 2021)

It has long been recognized that neural development relies intimately on activity-dependent mechanisms during early developmental windows known as "critical periods". We recently showed that experiences during critical periods are far more influential, and determine the spatial, functional, and circuit attributes of cortical neurons in the upper layers of the neocortex (Cheng et al., Cell, 2022). In contrast, vision seems to be less important for cell type specification in the retina (Whitney et al., Neuroscience, 2023).

image.png
image.png
image.png
image.png

How did neural diversity evolve?

The basic plan of the retina is beautifully conserved across all living vertebrates, despite species differing profoundly in their visual needs. Some species are diurnal, other nocturnal; some are terrestrial, others aquatic; some mainly hunt, others forage for colorful fruits. How neuron types in the retina might have evolved to accommodate these varied needs has not been systematically studied. Using large-scale single-cell transcriptomics, we have shown that despite extensive variation,  numerous retinal types are shared across species based on conserved gene expression programs that likely trace back to an ancestral vertebrate >400 million years ago. We have identified rodent orthologs of the so-called "midget" cells which comprise >80% of the retinal output in humans, subserve high acuity vision, and were hitherto believed to be specific to primates (Hahn et al., Nature, 2023). The orthology relationship is surprising, and provokes an intriguing hypothesis connecting the evolution of the retina to the evolution of the cerebral cortex. Knowing the orthologs of midget cells in several accessible models will aid efforts to slow their degeneration in blinding diseases such as glaucoma. Together with collaborators, we have successfully shown, as predicted by transcriptomics, the neural pathways that enable us to see in the dark likely arose in the common ancestor of all vertebrates (mammals, fish, reptiles and birds)! (Hellevik et al., 2023)

image.png
image.png

Spatiotemporal dynamics of diffuse charges near neuronal membranes

Neurons communicate with each other through an electrochemical mechanism called the action potential, which involves a transient change in membrane potential propagating throughout the neuron in an "all or none" fashion. We are interested in understanding fundamental aspects of action potential generation and propagation, particularly the role of ion transport, electrostatic, mechanical and osmotic effects, which have been neglected in classical "circuit" descriptions. We are combining continuum and statistical mechanical models and large-scale numerical simulations to address this fundamental question.
 

image.png
image.png
bottom of page