CausalPath works on a set of molecular profiles and predicts the causal mechanisms that acts on the data. It integrates prior knowledge from biological pathway databases, and uses them to explain new observations. While the main focus is phosphoproteomics, CausalPath can also work on global proteomic, transcriptomic, acetylomic and methylomic profiles, and combinations of them.

Utilized: HTML, JavaScript, CSS, Node JS, Git, Chrome Dev Tools

I've developed a comprehensive solution for single-cell pathway analysis:
1. Client: I built a user-friendly web app with React JS for generating, visualizing, and modifying cell trajectories.
2. Server: A robust Django-based server handles communication with real-time responses and data storage.
3. Database: I use MongoDB for data storage, ensuring data integrity.
I've also implemented a Python pipeline for efficient multi-step analysis, making pathway analysis more accessible to researchers. My goal is to advance bioinformatics and simplify complex analyses for a better understanding of cellular behavior.

Utilized: React JS, Django, MongoDB, HTML, Tailwind CSS, JavaScript, Rest API, Graph JS, Python, Git

Server GitHub: https://github.com/PathwayAndDataAnalysis/single-cell-pathway-analysis

Client GitHub: https://github.com/PathwayAndDataAnalysis/single-cell-pathway-analysis-client

Developed a new transcription factor activity prediction tool using Python, Numpy, and Pandas. I implemented parallelization in Python to work on large CSV files(more than 600 MB) and to run the analysis for each cell. This project is based on rank-based analysis.

Utilized: Python, Numpy, Pandas, Git, Jupyter Notebook, Google Colab, Vectorization, Parallelization

https://github.com/PathwayAndDataAnalysis/TF-Analysis

I built a machine-learning model to classify whether a gene is mutated or not based on its read count. As genes mutate so frequently, some mutations are harmful some are not. As we have many datasets already available about the gene mutations. So if we can detect gene is mutated or not of a patient using machine learning models, we can infer other information about the patient like suggesting a similar type of treatment to other patients who also have the same mutation. I used a dataset from The Cancer Genome Atlas (TCGA), a landmark cancer genomics program, molecularly characterized over 20,000 primary cancer and matched normal samples spanning 33 cancer types.

Utilized: Python, Tensorflow, Keras, Pandas, Numpy, Git, Jupyter Notebook, Google Colab

Source Code: https://github.com/KisanThapa/GeneMutationClassification

Project Report: https://sites.google.com/view/ml-project-report/home

It allows users to browse news feeds, friends' timelines, pages, and liked groups with an integrated browser. This app enables users to download and save videos from Facebook with a simple click, allowing for offline viewing or sharing with friends via various apps.I published it to Google Play Store.

1. "Protein Kinase Signaling Networks Driven by Oncogenic Gq/11 in Uveal Melanoma Identified by Phosphoproteomic and Bioinformatic Analyses" Michael D Onken, Petra Erdmann-Gilmore, Qiang Zhang, Kisan Thapa, Emily King, Kevin M Kaltenbronn, Sarah E Noda, Carol M Makepeace, Dennis Goldfarb, Özgün Babur, R Reid Townsend, Kendall J Blumer

   Molecular & Cellular Proteomics 22 (11)

   doi: https://doi.org/10.1016/j.mcpro.2023.100649
2. "A prognostic matrix code defines functional glioblastoma phenotypes and niches" Monika Vishnoi, Zeynep Dereli, Zheng Yin, Elisabeth K. Kong, Meric Kinali, Kisan Thapa, Ozgun Babur, Kyuson Yun, Nourhan Abdelfattah, Xubin Li, Behnaz Bozorgui, Robert C. Rostomily, Anil Korkut

   bioRxiv 2023.06.06.543903;

   doi: https://doi.org/10.1101/2023.06.06.543903
3. "A framework for considering prior information in network-based approaches to--omics data analysis" Julia Somers, Madeleine R Fenner, Dharani Thirumalaisamy, Garth Kong, William Yashar, Meric Kinali, Kisan Thapa, Olga Nikolova, Özgün Babur, Emek Demir

   Proteomics 23 (21-22), 2200402

   doi: https://doi.org/10.22541/au.168984016.69483308/v1
4. "Performance of chatgpt on usmle: Unlocking the potential of large language models for ai-assisted medical education" Prabin Sharma, Kisan Thapa, Prastab Dhakal, Mala Deep Upadhaya, Santosh Adhikari, Salik Ram Khanal

   arXiv:2307.00112 [cs.CY, cs.AI] 2023.

   doi: https://doi.org/10.48550/arXiv.2307.00112
5. "Screening Autism Spectrum Disorder in childrens using Deep Learning Approach : Evaluating the classification model of YOLOv8 by comparing with other models" Subash Gautam, Prabin Sharma, Kisan Thapa, Mala Deep Upadhaya, Dikshya Thapa, Salik Ram Khanal, Vítor Manuel de Jesus Filipe

   arXiv:2306.14300 [cs.CV, cs.AI] 2023.

   doi: https://doi.org/10.48550/arXiv.2306.14300
6. "Towards blind user's indoor navigation: a comparative study of beacons and decawave for indoor accurate location" Prabin Sharma, Sambad Bidari, Kisan Thapa, Antonio Valente, Hugo Paredes
   doi: https://doi.org/10.48550/arXiv.1912.01499