About Me
I'm Saiyang Na, a Computer Science Ph.D. candidate supervised by Dr. Junzhou Huang in Computer Science Department at University of Texas at Arlington, USA. I focus on creating open-source libraries with JAX and Python for deep learning and LLM applications.
Research
- Cell Segmentation - Auto-prompt generation pipeline for SAM
- Protein-Protein Interaction - TCR-pMHC binding prediction
- Deep Learning in Hyperbolic Space - MCI classification
- Retrosynthesis - Multi-step retrosynthesis on USPTO dataset
Projects
- nadl - A deep learning framework based on JAX and Equinox
- jaxrie - A JAX hyperbolic neural networks library
- naipyext - IPython Extensions with better trace exception
- Segment Any Cell Demo
Skills
Expert: Python, JAX with Equinox, PyTorch, NumPy
Proficient: TensorFlow with Keras, Lisp (Emacs Lisp), LLM
Familiar: Haskell, JavaScript, C/C++
Publications
- Saiyang Na et al., (2024), "Segment Any Cell: A SAM-based Auto-prompting Fine-tuning Framework for Nuclei Segmentation", TNNLS (Under review)
- Bing Song, Kaiwen Wang, Saiyang Na et al., (2024), "Cmai: Predicting Antigen-Antibody Interactions from Massive Sequencing Data", Nature Cancer (2nd round review).
- Bing Song, Kaiwen Wang, Saiyang Na et al., (2024), "An Artificial Intelligence Model for Profiling the Landscape of Antigen-binding Affinities of Massive BCR Sequencing Data", bioRxiv
- Feng Jiang, Yuzhi Guo, Hehuan Ma, Saiyang Na et. al, "GTE: a graph learning framework for prediction of T-cell receptors and epitopes binding specificity", Briefings in Bioinformatics
- Lu Zhang, Saiyang Na et al., (2023), "Multimodal deep fusion in hyperbolic space for mild cognitive impairment study", The 26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Oral.
- Xinyue Ye, Jiaxin Du, Xi Gong, Saiyang Na et al., (2021), "Geospatial and semantic mapping platform for massive COVID-19 scientific publication search", Journal of Geovisualization and Spatial Analysis