A5077534429- Mitochondrial Function and Pathology
- RNA Research and Splicing
- Genetic Neurodegenerative Diseases
- Parallel Computing and Optimization Techniques
- Cell Image Analysis Techniques
- Advanced Data Storage Technologies
- RNA and protein synthesis mechanisms
- Cloud Computing and Resource Management
- Electron and X-Ray Spectroscopy Techniques
- COVID-19 diagnosis using AI
- Advanced Electron Microscopy Techniques and Applications
- Genomics and Phylogenetic Studies
- Nuclear Physics and Applications
- Advanced X-ray Imaging Techniques
- Domain Adaptation and Few-Shot Learning
- Distributed and Parallel Computing Systems
- Computer Graphics and Visualization Techniques
- Anomaly Detection Techniques and Applications
- Enzyme Structure and Function
- RNA modifications and cancer
- Medical Imaging Techniques and Applications
- Multimodal Machine Learning Applications
- Mobile Crowdsensing and Crowdsourcing
- Machine Learning in Materials Science
- Embedded Systems Design Techniques
Stanford University
2022-2025
Birla Institute of Technology and Science, Pilani
2020
Birla Institute of Technology and Science, Pilani - Goa Campus
2017-2018
Tata Consultancy Services (India)
2018
Abstract Huntington’s disease (HD) is caused by an expanded CAG repeat in the huntingtin gene, yielding a Huntingtin protein with polyglutamine tract. While experiments patient-derived induced pluripotent stem cells (iPSCs) can help understand disease, defining pathological biomarkers remains challenging. Here, we used cryogenic electron tomography to visualize neurites HD patient iPSC-derived neurons varying repeats, and primary cortical from BACHD, deltaN17-BACHD, wild-type mice. In...
Combining three-dimensional (3D) visualization with elemental analysis of vitrified cells can provide crucial insights into subcellular structures and compositions in their native environments. We present a coordinated approach using cryogenic electron energy loss spectroscopy (cryoEELS) tomography (cryoET) to characterize the distribution ultrastructure cells. applied this method examine calcium disposition mitochondria cultured human retinal ganglion (RGCs) exposed pro-calcifying...
ABSTRACT Cryogenic electron tomography (cryoET) directly visualizes subcellular structures in 3D at the nanometer scale. Quantitative analyses of cryoET data can reveal structural biomarkers diseases, provide novel mechanistic insights, and inform effects treatments on phenotype. However, existing automated annotation approaches primarily focus localizing molecular features with few methods accurately quantifying complex such as organelles. We address this challenge CryoViT, a paradigm shift...
RNA protein interactions (RPI) play a pivotal role in the regulation of various biological processes. Experimental validation RPI has been time-consuming, paving way for computational prediction methods. The major limiting factor these methods accuracy and confidence predictions, our in-house experiments show that they fail to accurately predict involving short sequences such as TERRA RNA. Here, we present data-driven model using gradient boosting classifier. Amino acids nucleotides are...
Combining three-dimensional (3D) visualization with elemental analysis of vitrified cells can provide crucial insights into subcellular structures and compositions in their native environments. We present a coordinated approach using cryogenic electron energy loss spectroscopy (cryoEELS) tomography (cryoET) to characterize the distribution ultrastructure cells. applied this method examine calcium disposition mitochondria cultured human retinal ganglion (RGCs) exposed pro-calcifying...
A good design abstraction framework for high performance computing should provide a higher level programming that strikes balance between the and visibility over hardware so software developer can write portable without having to understand nuances, yet exploit compute power optimally. In this paper we have analyzed popular called "Thrust" from NVIDIA, proposed an extension Thrust++ provides memory hierarchy of NVIDIA GPU. allows developers make efficient use shared overall, better control...
High performance computing applications are far more difficult to write, therefore, practitioners expect a well-tuned software last long and provide optimized even when the hardware is upgraded. It may also be necessary write using sufficient abstraction over so that it capable of running on heterogeneous architecture. Therefore, required have proper programming paradigm strikes balance between visibility programmer can program without having understand nuances, yet exploit compute power...
Foundation vision or vision-language models are trained on large unlabeled noisy data and learn robust representations that can achieve impressive zero- few-shot performance diverse tasks. Given these properties, they a natural fit for active learning (AL), which aims to maximize labeling efficiency, but the full potential of foundation has not been explored in context AL, specifically low-budget regime. In this work, we evaluate how influence three critical components effective namely, 1)...
Recent advances in microscopy have enabled the rapid generation of terabytes image data cell biology and biomedical research. Vision-language models (VLMs) offer a promising solution for large-scale biological analysis, enhancing researchers' efficiency, identifying new biomarkers, accelerating hypothesis scientific discovery. However, there is lack standardized, diverse, vision-language benchmarks to evaluate VLMs' perception cognition capabilities understanding. To address this gap, we...
When crowdsourcing systems are used in combination with machine inference the real world, they benefit most when system is deeply integrated crowd workers. However, if researchers wish to integrate "off-the-shelf" classifiers, this deep integration not always possible. This work explores two strategies increase accuracy and decrease cost under setting. First, we show that reordering tasks presented human can create a significant improvement. Further, greedily choosing parameters maximize...
Abstract Huntington’s Disease (HD) is caused by an expanded CAG repeat in the huntingtin gene, yielding a Huntingtin protein with polyglutamine tract. Patient-derived induced pluripotent stem cells (iPSCs) can help understand disease; however, defining pathological biomarkers challenging. Here, we used cryogenic electron tomography to visualize neurites HD patient iPSC-derived neurons varying repeats, and primary cortical from BACHD, deltaN17-BACHD, wild-type mice. In models, discovered...
High performance computing applications are far more difficult to write, therefore, practitioners expect a well-tuned software last long and provide optimized even when the hardware is upgraded. It may also be necessary write using sufficient abstraction over so that it capable of running on heterogeneous architecture. A good design paradigm strikes balance between visibility hardware. This allows programmer without having understand nuances while exploiting power optimally. In this paper we...
Computational efforts to understand the process of pre-mRNA splicing have led development various tools that can separately predict locations branchpoints and splice sites. Since experimental studies shown majority are distributed just upstream 3' site, it should be possible use this information jointly branchpoint acceptor site. Here, we propose a deep neural network based sequence-to-sequence learning solution label each nucleotide an input sequence as being or site neither. We demonstrate...
Machine Learning (ML) methods have been used with varying degrees of success on protein prediction tasks, two inherent limitations. First, performance often depends upon the features extracted from proteins. Second, experimental data may be insufficient to construct reliable ML models. Here we introduce MP3vec, a transferable representation for sequences that is designed specifically sequence-to-sequence learning tasks. We use transfer generate MP3vecs by training deep neural network source...
Abstract Huntington’s Disease (HD) is caused by an expanded CAG repeat in the huntingtin gene, yielding a Huntingtin protein with polyglutamine tract. Patient-derived induced pluripotent stem cells (iPSCs) can help understand disease; however, defining pathological biomarkers challenging. Here, we used cryogenic electron tomography to visualize neurites HD patient iPSC-derived neurons varying repeats, and primary cortical from BACHD, deltaN17-BACHD, wild-type mice. In models, discovered...