Sanket Rajan Gupte

Sanket Rajan Gupte

Auto-generated profile, based on publicly available information

Sanket Rajan Gupte

Department of Computer Science, Stanford University

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
22 June 2018 in Scientific Reports
Verified
Journal Article
Open access
PDF file
DJDharm Skandh Jain
Dharm Skandh Jain
SGSanket Rajan Gupte
Sanket Rajan Gupte
RARaviprasad Aduri
Raviprasad Aduri
RNA protein interactions (RPI) play a pivotal role in the regulation of various biological processes. Experimental validation of RPI has been time-consuming, paving the way for computational prediction methods. The major limiting factor of these methods has been the accuracy and confidence of the predictions, and our in-house experiments show that they fail to accurately predict RPI involving short RNA sequences such as TERRA RNA. Here, we present a data-driven model for RPI prediction using a gradient boosting classifier. Amino acids and nucleotides are classified based on the high-resolution structural data of RNA protein complexes. The minimum structural unit consisting of five residues is used as the descriptor. Comparative analysis of existing methods shows the consistently higher performance of our method irrespective of the length of RNA present in the RPI. The method has been successfully applied to map RPI networks involving both long noncoding RNA as well as TERRA RNA. The method is also shown to successfully predict RNA and protein hubs present in RPI networks of four different organisms. The robustness of this method will provide a way for predicting RPI networks of yet unknown interactions for both long noncoding RNA and microRNA.

1 December 2017
Verified
Conference Paper
Closed access
AGAjai V. George
Ajai V. George
SMSankar Manoj
Sankar Manoj
SGSanket R. Gupte
Sanket R. Gupte
SMSayantan Mitra
Sayantan Mitra
SSSantonu Sarkar
Santonu Sarkar
A good design abstraction framework for high performance computing should provide a higher level programming abstraction that strikes a balance between the abstraction and visibility over the hardware so that the software developer can write a portable software without having to understand the hardware nuances, yet exploit the compute power optimally. In this paper we have analyzed a popular design abstraction framework called "Thrust" from NVIDIA, and proposed an extension called Thrust++ that provides abstraction over the memory hierarchy of an NVIDIA GPU. Thrust++ allows developers to make efficient use of shared memory and overall, provides better control over the GPU memory hierarchy while writing applications in Thrust style for the CUDA backend. We have shown that when applications are written for the CUDA backend using Thrust++, they have minimal performance degradation when compared to their equivalent CUDA versions. Further, Thrust++ provides almost 4x speedup when compared to Thrust, for certain compute intensive kernels that repeatedly use the reduce operation.

30 June 2024 in bioRxiv
Verified
Preprint
Open access
PDF file
SGSanket R. Gupte
Sanket R. Gupte
CHCathy Hou
Cathy Hou
GWGong-Her Wu
Gong-Her Wu
JGJesus G Galaz-Montoya
Jesus G Galaz-Montoya
WCWah Chiu
Wah Chiu
SYSerena Yeung-Levy
Serena Yeung-Levy
Cryogenic electron tomography (cryoET) directly visualizes subcellular structures in 3D at the nanometer scale. Quantitative analyses of cryoET data can reveal structural biomarkers of diseases, provide novel mechanistic insights, and inform the effects of treatments on phenotype. However, existing automated annotation approaches primarily focus on localizing molecular features with few methods accurately quantifying complex structures such as organelles. We address this challenge with CryoViT, a paradigm shift from traditional convolutional neural networks that leverages vision transformers to enhance the segmentation of large pleomorphic structures that can occupy almost the entire field of view in high-magnification images, such as mitochondria. CryoViT is powered by a large-scale vision foundation model and overcomes limitations of popular U-Net based methods, particularly when training data are scarce. We demonstrate the efficacy of CryoViT on a large cryoET dataset of neurons differentiated from iPSCs derived from Huntington disease (HD) patients and cultured HD mouse model neurons.

26 June 2017
Verified
Conference Paper
Open access
AGAjai V. George
Ajai V. George
SMSankar Manoj
Sankar Manoj
SGSanket R. Gupte
Sanket R. Gupte
SSSantonu Sarkar
Santonu Sarkar
High performance computing applications are far more difficult to write, therefore, practitioners expect a well-tuned software to last long and provide optimized performance even when the hardware is upgraded. It may also be necessary to write software using sufficient abstraction over the hardware so that it is capable of running on heterogeneous architecture. A good design abstraction paradigm strikes a balance between the abstraction and visibility over the hardware. This allows the programmer to write applications without having to understand the hardware nuances while exploiting the computing power optimally. In this paper we have analyzed the power of design abstraction of a popular design abstraction framework called Thrust both from ease of programming and performance perspectives. We have shown that while Thrust framework is good in describing an algorithm compared to the native CUDA or OpenMP version but it has quite a few design limitations. With respect to CUDA it does not provide any abstraction over the shared, texture or constant memory usage to the programmer. We have compared the performance of a Thrust application code in CUDA, OpenMP and the CPP backends with respect to the native versions (implementing exactly same algorithm), written for these backends and found that the current Thrust version performs poorly in most of the cases. While we conclude that the framework is not ready for writing applications that can exploit the optimal performance from the hardware, we also highlight the improvements necessary for the framework to make the performance comparable.

20 December 2024 in bioRxiv
Verified
Preprint
Open access
PDF file
GWGong-Her Wu
Gong-Her Wu
CHCathy Hou
Cathy Hou
ATAndrew Thron
Andrew Thron
HPHirenkumar Rajendra Patel
Hirenkumar Rajendra Patel
LSLiam Spillane
Liam Spillane
WCWah Chiu
Wah Chiu
Summary: Combining three-dimensional (3D) visualization with elemental analysis of vitrified cells can provide crucial insights into subcellular structures and elemental compositions in their native environments. We present a coordinated approach using cryogenic electron energy loss spectroscopy (cryoEELS) and cryogenic electron tomography (cryoET) to characterize the elemental distribution and ultrastructure of vitrified cells. We applied this method to examine calcium disposition in the mitochondria of cultured human retinal ganglion cells (RGCs) exposed to pro-calcifying conditions relevant to optic disc drusen pathology. Our cryoEELS analysis revealed mitochondrial granules with elevated calcium signals, offering direct evidence of mitochondrial calcification. Additionally, cryoET coupled with artificial intelligence-based analysis enabled quantification of the volume and spatial distribution of these calcium granules. This integrated workflow can be broadly applied to various cell types, facilitating the study of ultrastructure and elemental distribution in subcellular structures under diverse physiological and pathological conditions, as well as in response to therapeutic interventions.

1 August 2017
Verified
Conference Paper
Closed access
AGAjai V. George
Ajai V. George
SMSankar Manoj
Sankar Manoj
SGSanket Rajan Gupte
Sanket Rajan Gupte
SSSantonu Sarkar
Santonu Sarkar
High performance computing applications are far more difficult to write, therefore, practitioners expect a well-tuned software to last long and provide optimized performance even when the hardware is upgraded. It may also be necessary to write software using sufficient abstraction over the hardware so that it is capable of running on heterogeneous architecture. Therefore, it is required to have a proper programming abstraction paradigm that strikes a balance between the abstraction and visibility over the hardware so that the programmer can write a program without having to understand the hardware nuances, yet exploit the compute power optimally. In this paper we have analyzed the power of design abstraction and performance of a popular design abstraction framework called Thrust. We have shown quantitatively that while it is easier to write an application using Thrust compared to writing the same in the native CUDA or OpenMP backends, the framework does not provide any abstraction over the memory hierarchy of the underlying backend to the programmer. We have compared the performance of three Thrust applications with their corresponding native versions in CUDA, OpenMP, Xeon-Phi and the CPP backends and demonstrate that the current Thrust version performs poorly in most of the cases when the application is compute intensive. However, the framework provides close to the native performance for a non-compute intensive applications. We analyze the reasons for the performance and highlight the improvements necessary for the framework.