Research

Our group focuses on the development and application of computational and experimental research methods geared toward human well-being. We combine expertise as biochemists with software development skills, and we are eager to use this training in overcoming obstacles and bottlenecks in our understanding of how biomolecules function through interaction with their microenvironment. We envision contributing indispensable high-throughput tools for analyzing biochemical/biophysical data and making technological breakthroughs towards a better understanding of protein 3D structures in terms of their roles and behaviors in biological context.

Our research operates across five interconnected themes, all unified by the POKY suite — our integrated, AI-powered NMR platform developed and maintained in this lab.

2. AI and LLM Integration in NMR and Structural Biology

Artificial intelligence is not merely accelerating structural biology — it is redefining what the discipline can ask. Our group works at this frontier, building the computational and reasoning infrastructure that lets AI move from a passive analysis tool into an active participant in NMR-based structural discovery.

Current directions include:

• From pattern recognition to scientific reasoning. We are developing AI systems that progress along a deliberate arc — from assistant (automating individual analysis steps) to copilot (interactively guiding experimental decisions) to agent (autonomously reasoning over competing structural hypotheses and determining what to examine next). This trajectory underlies every project below.

• Deep learning for NMR peak detection and assignment. Convolutional and transformer-based architectures automate the recognition of spectral patterns that have historically required years of expert training to interpret reliably.

• LLM-assisted structural reasoning pipelines. Large language models are integrated directly into the experimental workflow — interpreting NMR data in context, assisting with experimental design, and generating structure reports whose logic remains fully traceable back to the underlying evidence.

• AI-integrated structure determination. We combine AI-based structure prediction (AlphaFold, Boltz-2, Chai-1, and related tools) with experimental NMR constraints to determine and validate structures of challenging targets, including homomeric protein complexes — the focus of our HANA algorithm.

• Chemical shift prediction as a native capability. We are building chemical shift predictors directly into automated NMR assignment pipelines, rather than treating prediction as an external, disconnected step.

This work sits deliberately at the intersection of structural biology, bioinformatics, and computer science, and is one of our most active recruiting areas for students drawn to programming and AI as much as to chemistry.

3. Integrative Structural Biology and Biomolecular Function

Understanding how proteins accomplish biological work requires knowing not just their three-dimensional structure but also their dynamics, interactions, and context within the cell. We approach this using an integrative structural biology strategy that combines NMR spectroscopy with complementary experimental and computational methods.

• NMR solution and solid-state structure determination: We determine high-resolution protein structures using both solution and solid-state NMR, with automated workflows enabled by POKY. Solid-state NMR expands our reach to membrane proteins, fibrils, and large assemblies that are inaccessible to solution methods.

• NMR/SAXS hybrid approaches: We combine solution NMR with small-angle X-ray scattering (SAXS) data for integrative structure determination of challenging targets. Our NMR/SAXS structure of the mitochondrial-targeted GTPase-activating protein VopE is a recent example (Smith & Lee et al., Protein Science 2022).

• Protein allostery: Allosteric regulation — where binding at one site changes protein behavior at a distant site — is a major mechanism of cellular control. We develop and apply CHESCA-based NMR methods, now fully automated within POKY (CHESPA/CHESCA-SPARKY), to map long-range allosteric networks in signaling proteins. Collaborative work with the Melacini group (McMaster University) has characterized allosteric mechanisms in PKA regulatory subunits and EPAC1. Recent work extended this to Platelet Factor 4 (PF4) tetramers (Ma et al., Biomol. NMR Assignments 2025).

• Protein–RNA interactions and infectious disease: We study protein complexes involved in RNA regulation and viral biology. Recent work characterized how SARS-CoV-2 Nsp9 depletes cellular microRNA let-7b (Mun et al., RNA Biology 2025). Ongoing projects investigate HRV-C (Human Rhinovirus C) components — including the 2Apro protease and CDHR3 receptor-binding domains — as targets for antiviral development. The only solution-state NMR structure of HRV-C2 2Apro was determined in our group.

• Cryo-EM: We have access to cryo-EM instrumentation through the core facility, and computational pipelines for integrating cryo-EM maps with NMR data are part of our expanding integrative toolkit.

4. NMR-Based Metabolomics

Metabolomics — the comprehensive profiling of small-molecule metabolites in biological systems — offers a powerful window into cellular physiology, disease states, and the effects of environmental or pharmacological perturbations. NMR is uniquely suited to metabolomics because it is quantitative, non-destructive, and requires minimal sample preparation.

Our group develops and applies computational tools that make NMR-based metabolomics more accessible and powerful (Chiu et al., Metabolomics 2024):

• A-SIMA (Advanced Support for Interactive Metabolite Analysis): A POKY-integrated tool for automated metabolite identification from 1D and 2D NMR spectra, featuring an intuitive graphical interface and complete user control over the analysis workflow.

• A-MAP (Automated Metabolomics Analysis and Processing): A multivariate statistical analysis tool supporting PCA and OPLS-DA, which takes NMR regions of interest as input to identify metabolic differences between sample groups.

Both tools are pre-built into the POKY suite with accompanying tutorial videos, substantially lowering the barrier to entry for new users and non-expert labs.

Lignin NMR: We have an active project developing standard NMR analysis methods and reference tools for lignin — a complex aromatic biopolymer and major component of plant biomass. Accurate NMR characterization of lignin structure is essential for biomass valorization, biofuel development, and sustainable materials chemistry. This project connects our NMR methodology expertise to important challenges at the interface of chemistry and environmental sustainability.

5. Computer-Aided Drug Design and Traditional Herbal Medicine

Structural data, when combined with computational docking, offers a powerful path to drug discovery — particularly for targets where experimental high-throughput screening is costly or slow. Our group integrates NMR-derived structural information with in-silico methods through ANSER/ADOCK, a multidimensional drug design module within the POKY suite.

ANSER/ADOCK enables:

• Virtual screening of compound libraries against NMR-determined protein structures

• Binding pose prediction and ranking using integrated scoring functions

• Iterative NMR validation of docking predictions through experimental ligand binding assays (e.g., chemical shift perturbation, STD-NMR)

Traditional Herbal Medicine (THM) and cancer drug discovery: One of our most distinctive and exciting research directions applies this CADD pipeline to the systematic, science-based investigation of traditional herbal medicine compounds. Herbal formulas used in East Asian medicine represent thousands of years of empirical pharmacological experience — yet the molecular mechanisms behind many of their effects remain poorly understood.

We screen curated herbal medicine compound libraries against cancer-relevant target proteins using ADOCK, identify promising hits, and validate binding using NMR spectroscopy. This approach bridges traditional knowledge and modern structural biology to accelerate identification of novel anticancer lead compounds.

A recent collaborative study demonstrated that a herbal formula derived from Danggui Bohyul Tang enhances cisplatin efficacy in lung cancer by inhibiting neutrophil extracellular trap (NET) formation — with our NMR structural biology contributing to mechanistic understanding (Kim et al., in revision). This work reflects our ongoing collaboration with colleagues in integrative medicine and oncology.

This research direction sits at the intersection of structural biology, pharmacology, and traditional medicine, and is particularly suited for students interested in translational or interdisciplinary science.

Research Infrastructure

Our lab is equipped to carry out the full spectrum of structural biology research from molecular biology bench to computational analysis:

Wet Lab / Biochemical Infrastructure

• Protein Expression & Cell Culture: Full capabilities for recombinant protein production (bacterial and insect expression systems, BSL-1) supported by temperature-controlled shaking incubators and autoclaves.

• Purification & Processing: Automated ÄKTA start protein purification systems for rapid separation and isolation.

• Characterization & Biophysics: Isothermal Titration Calorimetry (ITC) for binding affinity analysis and NanoDrop spectrophotometer for precise micro-volume quantitation.

Dry Lab / Computing

• AI & Deep Learning: In-house NVIDIA A100 GPU cluster, high-CPU servers, and three NVIDIA DGX SPARK for advanced model training

• Storage & Data Management: Synology Enterprise NAS Server System

• High-Performance Computing: Full access to a scalable HPC cluster environment

• Development Pipeline: Comprehensive POKY suite development and testing environment

NMR

• Access to 400–900 MHz Bruker NMR spectrometers

• Solution and solid-state NMR capabilities

Structural Imaging

• Cryo-EM access through core facility

Visit "Publications" section on top to read related publications.

Watch YouTube videos from Woonghee's channel to learn software tools that we develop. Subscribe and turn notification on!

www.youtube.com/user/pinesparky

SI 0010 NMR Schedule