Bishal Upadhyaya

Bishal Upadhyaya

Auto-generated profile, based on publicly available information

Bishal Upadhyaya

Thermo Fisher Scientific UC San Diego NVIDIA Corporation

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 October 2019 in PLOS Genetics
Verified
Journal Article
Open access
PDF file
LVLisa Voelker
Lisa Voelker
BUBishal Upadhyaya
Bishal Upadhyaya
DFDenise M Ferkey
Denise M Ferkey
SWSarah Woldemariam
Sarah Woldemariam
NLNoelle D L'Etoile
Noelle D L'Etoile
JBJihong Bai
Jihong Bai
In order to respond to changing environments and fluctuations in internal states, animals adjust their behavior through diverse neuromodulatory mechanisms. In this study we show that electrical synapses between the ASH primary quinine-detecting sensory neurons and the neighboring ASK neurons are required for modulating the aversive response to the bitter tastant quinine in C. elegans. Mutant worms that lack the electrical synapse proteins INX-18 and INX-19 become hypersensitive to dilute quinine. Cell-specific rescue experiments indicate that inx-18 operates in ASK while inx-19 is required in both ASK and ASH for proper quinine sensitivity. Imaging analyses find that INX-19 in ASK and ASH localizes to the same regions in the nerve ring, suggesting that both sides of ASK-ASH electrical synapses contain INX-19. While inx-18 and inx-19 mutant animals have a similar behavioral phenotype, several lines of evidence suggest the proteins encoded by these genes play different roles in modulating the aversive quinine response. First, INX-18 and INX-19 localize to different regions of the nerve ring, indicating that they are not present in the same synapses. Second, removing inx-18 disrupts the distribution of INX-19, while removing inx-19 does not alter INX-18 localization. Finally, by using a fluorescent cGMP reporter, we find that INX-18 and INX-19 have distinct roles in establishing cGMP levels in ASK and ASH. Together, these results demonstrate that electrical synapses containing INX-18 and INX-19 facilitate modulation of ASH nociceptive signaling. Our findings support the idea that a network of electrical synapses mediates cGMP exchange between neurons, enabling modulation of sensory responses and behavior. Animals are constantly adjusting their behavior to respond to changes in the environment or to their internal state. This behavior modulation is achieved by altering the activity of neurons and circuits through a variety of neuroplasticity mechanisms. Chemical synapses are known to impact neuroplasticity in several different ways, but the diversity of mechanisms by which electrical synapses contribute is still being investigated. Electrical synapses are specialized sites of connection between neurons where ions and small signaling molecules can pass directly from one cell to the next. By passing small molecules through electrical synapses, neurons may be able to modify the activity of their neighbors. In this study we identify two genes that contribute to electrical synapses between two sensory neurons in C. elegans. We show that these electrical synapses are crucial for proper modulation of sensory responses, as without them animals are overly responsive to an aversive stimulus. In addition to pinpointing their sites of action, we present evidence that they may be contributing to neuromodulation by facilitating passage of the small molecule cGMP between neurons. Our work provides evidence for a role of electrical synapses in regulating animal behavior.

5 August 2019 in bioRxiv
Verified
Preprint
Open access
LVLisa Voelker
Lisa Voelker
BUBishal Upadhyaya
Bishal Upadhyaya
DFDenise M. Ferkey
Denise M. Ferkey
SWSarah Woldemariam
Sarah Woldemariam
NLNoelle D. L’etoile
Noelle D. L’etoile
JBJihong Bai
Jihong Bai
In order to respond to changing environments and fluctuations in internal states, animals adjust their behavior through diverse neuromodulatory mechanisms. In this study we show that electrical synapses between the ASH primary quinine-detecting sensory neurons and the neighboring ASK neurons are required for modulating the aversive response to the bitter tastant quinine in C. elegans . Mutant worms that lack the electrical synapse proteins INX-18 and INX-19 become hypersensitive to dilute quinine. Cell-specific rescue experiments indicate that inx-18 operates in ASK while inx-19 is required in both ASK and ASH for proper quinine sensitivity. Imaging analyses find that INX-19 in ASK and ASH localizes to the same regions in the nerve ring, suggesting that both sides of ASK-ASH electrical synapses contain INX-19. While inx - 18 and inx - 19 mutant animals have a similar behavioral phenotype, several lines of evidence suggest the proteins encoded by these genes play different roles in modulating the aversive quinine response. First, INX-18 and INX-19 localize to different regions of the nerve ring, indicating that they are not present in the same synapses. Second, removing inx-18 disrupts the distribution of INX-19, while removing inx-19 does not alter INX-18 localization. Finally, by using a fluorescent cGMP reporter, we find that INX-18 and INX-19 have distinct roles in establishing cGMP levels in ASK and ASH. Together, these results demonstrate that electrical synapses containing INX-18 and INX-19 facilitate modulation of ASH nociceptive signaling. Our findings support the idea that a network of electrical synapses mediates cGMP exchange between neurons, enabling modulation of sensory responses and behavior. Author Summary: Animals are constantly adjusting their behavior to respond to changes in the environment or to their internal state. This behavior modulation is achieved by altering the activity of neurons and circuits through a variety of neuroplasticity mechanisms. Chemical synapses are known to impact neuroplasticity in several different ways, but the diversity of mechanisms by which electrical synapses contribute is still being investigated. Electrical synapses are specialized sites of connection between neurons where ions and small signaling molecules can pass directly from one cell to the next. By passing small molecules through electrical synapses, neurons may be able to modify the activity of their neighbors. In this study we identify two genes that contribute to electrical synapses between two sensory neurons in C. elegans . We show that these electrical synapses are crucial for proper modulation of sensory responses, as without them animals are overly responsive to an aversive stimulus. In addition to pinpointing their sites of action, we present evidence that they may be contributing to neuromodulation by facilitating passage of the small molecule cGMP between neurons. Our work provides evidence for a role of electrical synapses in regulating animal behavior.

28 August 2023 in arXiv
Verified
Preprint
Open access
PDF file
BLBrad Larson
Brad Larson
BUBishal Upadhyaya
Bishal Upadhyaya
LMLuke McDermott
Luke McDermott
SGSiddha Ganju
Siddha Ganju
Deep learning harnesses massive parallel floating-point processing to train and evaluate large neural networks. Trends indicate that deeper and larger neural networks with an increasing number of parameters achieve higher accuracy than smaller neural networks. This performance improvement, which often requires heavy compute for both training and evaluation, eventually needs to translate well to resource-constrained hardware for practical value. Structured pruning asserts that while large networks enable us to find solutions to complex computer vision problems, a smaller, computationally efficient sub-network can be derived from the large neural network that retains model accuracy but significantly improves computational efficiency. We generalize structured pruning with algorithms for network augmentation, pruning, sub-network collapse and removal. In addition, we demonstrate efficient and stable convergence up to 93% sparsity and 95% FLOPs reduction without loss of inference accuracy using with continuous relaxation matching or exceeding the state of the art for all structured pruning methods. The resulting CNN executes efficiently on GPU hardware without computationally expensive sparse matrix operations. We achieve this with routine automatable operations on classification and segmentation problems using CIFAR-10, ImageNet, and CityScapes datasets with the ResNet and U-NET network architectures.