Welcome to my Website

Welcome to my Website

Umesh Khaniya

Hi, I'm Umesh Khaniya

I am a postdoctoral researcher at NIH working on antibody engineering, CAR-T cell modeling, and computational immunology. My focus includes structure-based analysis of Ig folds, machine learning–based topology labeling, and predicting domain-domain interactions.

Education

  • Ph.D. in Physics, CUNY Graduate Center, 2016–2022
  • Master in Physics, Tribhuvan University, Nepal, 2013–2015

🔬 Key Skills

  • MD & Docking: NAMD, GROMACS, OpenMM, CHARMM-GUI, MM-PBSA, FEP, Schrödinger BioLuminate, AutoDock, PIPER
  • Protein Modeling: AlphaFold (AF2/AF3), ESMFold, RoseTTAFold, Modeller, Chai Discovery
  • ML: Graph Neural Networks, Transformer Models, Diffusion Models, Hugging Face, Fine-Tuning
  • Frameworks: PyTorch, TensorFlow, scikit-learn, PySpark
  • Protein-Ligand Docking: Schrödinger BioLuminate, PIPER, AutoDock
  • Cheminformatics: RDKit, PaDEL
  • Programming: Python, SQL, Bash, R
  • Cloud & DevOps: AWS (EC2, S3, Redshift, Lambda), HPC environments, Docker, Git, Airflow
  • Visualization & Tools: VMD, PyMol, UCSF Chimera, Jupyter Notebook

Bio

I am a postdoctoral researcher at NIH focused on antibody engineering, CAR-T modeling, and machine learning-based structural analysis.

Papers

  • Paper 1: Title and link
  • Paper 2: Title and link

Experience

  • Postdoc, NIH (2022–Present)
  • Graduate Researcher, CUNY (2016–2022)

Hobby

Hiking, photography, and reading about AI and science history.

Projects

  • IgStrand universal numbering
  • CAR-T structure prediction

📂 Projects

  • IgStrand Universal Numbering: Structural classification of Ig domains across proteomes using TM-align and AF2 models.
  • CAR-T Modeling: Developed pipelines to simulate and evaluate synthetic CAR-T constructs using MD simulations and structural prediction tools.

Download CV