Smith Gupta

Lab 14, Biological Sciences and Bioengineering,
Indian Institute of Technology Kanpur,
Kanpur, India - 208016.
smith [at] iitk [dot] ac [dot] in
Generic placeholder image

I am a research fellow at the Biological Sciences and Bioengineering department at IIT Kanpur. I am working with Dr. Nitin Gupta at the Lab of Neural Systems in the area of insect olfaction. My research aims to decipher the mechanism by which neuromodulators overcome the variability of the nervous system to generate reliable behavior. We have already quantified the variation in the antennal lobe of Aedes aegypti and now want to understand the compensatory mechanisms that act during animal behavior. Towards this goal, we have created an arena that can simulate an olfactory environment of any complexity, allowing a fly to freely navigate in the environment while we simultaneously record calcium signals from neurons of interest.

Education


Indian Institute of Technology, Kanpur

Jan 2019 - present
Ph.D. in Biological Sciences

Indian Institute of Technology, Kanpur

July 2011 - May 2015
B.Tech. in Electrical Engineering

Publications


Preprint

An algorithm for detection of small-amplitude spikes in intracellular recordings. Smith Gupta, Toshi Hige, Nitin Gupta. [aRxiv]

Conferences

Understanding olfactory processing in the antennal lobe in Aedes aegypti. Pranjul Singh, Shefali Goyal, Smith Gupta, Nitin Gupta. SFN, 2021.

Responses of the antennal lobe neurons to carbon dioxide in mosquitoes. Shefali Goyal, Pranjul Singh, Smith Gupta, Nitin Gupta. SFN, 2021.

Diversity of odor responses in maxillary palps of Aedes aegypti. Swikriti Singh, Abhishek Airan, Smith Gupta, Nitin Gupta. SFN, 2021.


Research Projects


Virtual Olfactory Arena


Variation in the antennal lobe of Aedes aegypti

The same olfactory MD3 projection neuron has different responses in different organisms; recordings below are from Aedes aegypti. Voltage traces shown are responses towards lactic acid, and the rasters show the responses of the same neurons towards a panel of 20 odorants. Same neurons across different organisms(sister PNs, shown in red violin) have only a weak correlation between their odor responses. We have quantified the variation at the antennal lobe level, which leaves us with the question concerning the compensatory mechanisms that result in consistent behavior towards these odorants, despite the variation.

We have also made methods to classify neuronal types based on their electrophysiological properties. Shown here is a panel of local neurons(above) and projection neurons(below) responding to a set of odorants. We parse features like spike amplitude, adaptation, accommodation, etc., which have a high correlation for the same cell types as shown in the heatmap.

How do odors look like in the head of an organism?

Here we have taken recordings from projection neurons of Aedes aegypti in response to a set of odorants. In each panel, a row represents the response of a neuron belonging to a distinct glomerulus, and neighboring rows represent proximate glomeruli. This visualization serves as a picture of odorants in a mosquito's brain.


An algorithm to detect small amplitude spikes

The typical amplitude of an action potential in vertebrate neurons is around 100 mV. However, in insects, because of the unipolar morphology and other structural characteristics, the sizes of the spikes recorded from the soma can be much smaller. Whole-cell patch-clamp recordings from the somata of projection neurons of the antennal lobe in mosquitoes can show spikes with amplitudes as small as 2 mV. Moreover, the observed spikes often ride on relatively large depolarizations, which makes it difficult for the standard thresholding-based approaches to distinguish them from noise or sharp EPSPs present in the signal.

For spike detection in such neuronal recordings, we propose a clustering-based algorithm that separates peaks corresponding to action potentials from those corresponding to noise. Candidate peaks, including many noise peaks, are first selected according to their sharpness, and then a feature vector is extracted for each peak. The 3-dimensional feature vector contains the absolute value of the peak voltage, height of the spike, and the magnitude of the second derivative minima attained during the spike.

In most recordings, this 3D space reveals two natural clusters, separating the noise peaks from the true action potentials.

 

In summary, the algorithm facilitates accurate spike detection to enable the interpretation and analysis of patch-clamp data from neuronal recordings in invertebrates. The algorithm is implemented in an open-source tool, which is freely available to the community. (Link)


Courses


Neurobiology

Methods and Tools in Cognitive Science

Cellular Molecular Biology

Computational Cognitive Science

Human Molecular Genetics

Algorithmic Information Theory


Find complete CV here.