🎓 Ph.D. Student in Computer Science 📍 University of Massachusetts Boston 💡 Researching Artificial Intelligence for Structural Bioinformatics and Drug Discovery

My work lies at the intersection of AI, structural biology, and computational modeling, with an emphasis on:

• Protein–protein and protein–ligand interactions
• Conformational dynamics and flexibility modeling
• AI-driven docking and drug repurposing
• Integration of sequence, structure, and biophysical features

I develop machine learning frameworks that capture the dynamic behavior of proteins and leverage these insights to improve binding prediction and molecular design. My research draws from deep learning, biophysics, and data-driven modeling to bridge the gap between computational predictions and experimental biology.

Development of transformer-based architectures for predicting protein–protein and protein–ligand interactions. Integrates sequence embeddings, SASA values, and rigidity-aware features using models such as ProtBERT, ESM-2, and Llama 3.

A Python toolkit for protein structure preprocessing, metadata annotation, and PDB management. Includes a graphical interface (PyQt5) and APIs for seamless integration with structural datasets.

Collaborative project integrating TM-align, AlphaFold 3, and AutoDock Vina to identify cross-species drug binding. Explores transferability of known drugs between human and Plasmodium homologous proteins.

A structured evaluation and benchmarking framework for Protein Language Models (pLMs). Scores models based on adaptability, bioinformatics relevance, and computational efficiency.

• Programming Languages: Python, C, Java, SQL, JavaScript

• Frameworks & Tools: PyTorch, TensorFlow, Hugging Face Transformers, NumPy, FAISS, AWS

• Specializations:

  • Deep learning model development for biological data
  • Structural bioinformatics and conformational analysis
  • Data integration from RCSB, ChEMBL, UniProt, and AlphaFold
  • Backend and full-stack system design for AI applications

• Instructor: Discrete Mathematics
• Teaching Assistant: Analysis of Algorithms and C Programming
• Student Advisory Council Member: Paul English Applied AI Institute, UMass Boston

• Advanced modeling of protein flexibility and binding mechanisms
• Hybrid AI–physics approaches for flexible docking
• Multimodal representation learning combining structure, sequence, and dynamics

I’m open to collaborations in AI for biology, drug discovery, and computational structural analysis.

📫 Email: k.ensafitakaldani001@umb.edu 🔗 LinkedIn: linkedin.com/in/kattayun-ensafi-370a20237 💻 GitHub: github.com/kattens

Outside of research, I enjoy creative media projects and video editing — having previously collaborated with a Persian YouTuber before transitioning fully into AI and computational biology.