PhD Scholar in Biomedical Engineering
📍 Syracuse, New York, USA
Highly motivated and result-oriented Researcher with a deep passion for integrating Computer science and computational biology to advance the understanding of life sciences field. Committed to utilizing these cutting-edge technologies to unlock a deeper understanding of biological systems, ultimately paving the way for transformative advancements in the fight against human diseases.
My research focuses on developing reproducible computational pipelines, interpretable models, and translating computational methods into biological insights.
Previously, I worked as a Research Associate at Aragen Life Sciences, building ML models for pharmacokinetic parameters and contributing to computer-aided drug design projects. I hold a Master's in Bioinformatics from Central University of Punjab, where my thesis focused on identifying blood transcriptome biomarkers for early Parkinson's disease diagnosis.
Specializing in scRNA-seq analysis with Seurat & Scanpy, deep learning with PyTorch & TensorFlow, and drug discovery with RDKit.
Conducting research in Single-cell RNA sequencing, Bioinformatics, and Biomedical Engineering.
Under the supervision of Dr. Samiron Phukan
Built ML models for pharmacokinetic parameters (Vdss, PPB) prediction.
Implemented deep learning solutions for small molecule generation using reinforcement learning.
Contributed to CADD research projects.
Under the supervision of Dr. Arti Sharma
Identified blood transcriptome biomarkers for early Parkinson's Disease diagnosis.
Achieved 90% accuracy with MLP algorithm on gene expression data from GEO database.
This repository contains an end-to-end pipeline for Single-Cell RNA Sequencing (scRNA-seq) data analysis using the Python library, Scanpy. It demonstrates rigorous Quality Control (QC), normalization, and dimensionality reduction (PCA/UMAP) on the public PBMC 3k dataset. The core output is the unbiased clustering (Leiden algorithm) to identify 6 distinct cell types, followed by Marker Gene Identification.
Using DL & ML algorithms our study provides a deeper and novel insights into the role of molecular descriptors in determining the PK parameters such as Vdss and PPB. FDA approved drugs with oral route of administration and having reported PK parameters were taken as the dataset. This was used for establishment of the foundational datasets followed by computation of different molecular descriptor classes.
Under the supervision of Dr. Samiron Phukan
Identified blood-based biomarkers for early PD diagnosis using machine learning on transcriptome data. MLP model achieved 90% accuracy differentiating patients from healthy controls.
Authors: Bantu, Rakesh; Phukan, Samiron; Haydar, Simon
Read Full Paper →A comprehensive guide to the free online resources, courses, and platforms that helped me master machine learning and deep learning. From beginner tutorials to advanced implementations, discover the learning path that can accelerate your AI journey without breaking the bank.
Read Article →A detailed roadmap for aspiring PhD students covering everything from choosing the right program to crafting compelling applications. Learn about essential documents, timeline management, funding opportunities, and insider tips that can make your application stand out in competitive programs.
Read Article →