THE MIN LAB | UNIST | 민두영 교수 연구실
watching membrane proteins fold, move, and interact — one molecule at a time
# biophysics, biochemistry, bioelectronics # AI-driven analysis, structural/bioinformatic analysis
# single-molecule experiments, protein/lipid chemistry, MD simulations
막단백질은 세포가 외부 신호를 감지하고, 서로 소통하며, 물질을 실어 나르는 과정을 좌우하며, 오늘날 의약품 표적의 대부분을 차지합니다. 그러나 막단백질의 일생에서 가장 중요한 사건들 — 어떻게 올바른 구조로 접히는지, 어떻게 서로 결합해 복합체를 이루는지, 시간이 지나며 어떻게 손상되는지 — 은 모두 복잡하고 기름진 세포막 환경 속에서 일어납니다. AI가 단백질의 최종 구조를 예측하는 시대가 되었음에도, 이 과정들은 여전히 충분히 밝혀지지 않았습니다. 막단백질의 이러한 숨겨진 사건들을 이해하는 일은 기초 생명과학인 동시에 인간 건강과도 깊이 맞닿은 질문입니다.
우리 연구실은 서로 맞물린 세 가지 질문을 탐구합니다. 막단백질이 세포막 안에서 어떻게 접히는가, 어떻게 상호작용하며 조립되어 기능성 복합체를 이루는가, 그리고 질병과 노화를 유발하는 분자적 손상을 어떻게 겪는가입니다. 우리는 이 질문들에 답하기 위해 단일 분자를 하나씩 직접 관찰합니다. 연구의 중심에는 자기집게를 이용한 단일분자 힘-형광 분광법이 있으며, 여기에 단백질·지질 생화학 및 재구성, 바이오전자 측정을 결합하고, 분자동역학 시뮬레이션과 AI 기반 구조·생명정보학 분석으로 이를 해석합니다. 이를 통해 지질 이중층 안에서 실제로 일어나는 분자적 사건을 직접 포착하고, 평균화된 측정으로는 볼 수 없는 원리를 밝혀냅니다.
아직 풀리지 않은 과학적 질문에 호기심을 갖고, 함께 답을 찾아갈 여러분을 기다리고 있습니다. 우리 연구는 매우 융합적인 성격을 띠고 있기 때문에, 화학, 생명과학, 물리학, 계산과학, 공학 등 다양한 배경을 가진 학생들을 환영합니다. 관심이 있다면 연구책임자인 민두영 교수(dymin@unist.ac.kr)에게 이메일을 보내주세요.
Membrane proteins govern how cells sense, communicate, and transport, and they account for the majority of today's drug targets. Yet the key events of their lifetime — how they fold into shape, how they assemble with one another, and how they are damaged over time — all unfold inside the crowded, oily world of the cell membrane, a process still poorly understood even in an era when AI can predict a protein's final structure. Understanding this hidden life of membrane proteins is a question of both fundamental life sciences and human health.
Our laboratory explores three linked questions: how membrane proteins fold within the membrane, how they interact and assemble into functional complexes, and how they suffer molecular damage that drives disease and aging. To address these questions, we watch single molecules one at a time. At the core of our approach is single-molecule force and fluorescence spectroscopy using magnetic tweezers. We combine this with protein/lipid biochemistry, reconstitution, and bioelectronic measurements, and interpret the results through molecular dynamics simulations and AI-driven structural and bioinformatic analysis. This lets us resolve events as they actually happen in a lipid bilayer, and reveal rules that averaged measurements cannot capture.
We are looking for curious minds excited to explore scientific questions that no one has answered yet. Because our work is highly interdisciplinary, we welcome students from diverse backgrounds, including chemistry, life sciences, physics, computational science, and engineering. If you’re interested, feel free to email the principal investigator, Prof. Duyoung Min, at dymin@unist.ac.kr.
-
Membrane Protein Folding, Misfolding, and Aggregation
-
Single-Molecule Magnetic Tweezers
Chemical Reviews 126, 6680 (2026) [LINK]
eLife 12, e85882 (2023) [LINK]
Science 366, 1150 (2019) [LINK]
Nature Chemical Biology 14, 489 (2018) [LINK]
Nature Chemical Biology 11, 981 (2015) [LINK]
Proteins must fold into precise three-dimensional shapes to work, and how they do so is one of the deepest questions in life sciences. For membrane proteins it is even harder: folding happens inside the oily, crowded environment of a cell membrane, a world we still understand poorly and one that AI structure prediction cannot yet explain. These proteins account for the majority of drug targets, so understanding how they fold matters not only for basic science but for medicine. We watch single molecules fold in real time, one at a time, to uncover rules that averaged measurements can never reveal.
In a living cell, membrane proteins almost never fold alone. Helper proteins called chaperones guide them, and when this guidance fails, misfolding and disease can follow. Yet how chaperones actually work has been remarkably hard to see, because it means catching a helper and its client in the act. This is where our single-molecule tools open a door that other methods cannot. Understanding how the cell actively shapes folding, rather than leaving it to chance, is a frontier that connects protein folding to quality control, aging, and disease.
-
Chaperone-Mediated Folding
-
Membrane Protein Interaction
-
Single-Molecule Interaction Tweezers
Nature Communications 16, 7366 (2025) [LINK]
Proteins rarely act in isolation. Many must find and bind their partners inside the membrane to carry out their jobs, from sensing signals to transporting molecules. How these partnerships form has long been treated as a simple on/off event, but the real process is far richer and mostly hidden from view. We developed a way to take a single protein pair apart and watch it come back together, step by step, inside a membrane. Revealing how molecular partnerships assemble helps explain how cells build the machinery of life, and how that machinery goes wrong in disease.
-
Molecular Dynamics Simulation
Nature Communications 16, 7366 (2025) [LINK]
Nano Today 65, 102834 (2025) [LINK]
Nature Communications 15, 10873 (2024) [LINK]
Nature Chemical Biology 14, 489 (2018) [LINK]
Experiments show us what a molecule does; simulations help us understand why. Molecular dynamics lets us follow every atom of a membrane protein as it moves, folds, and interacts. By pairing these simulations with our single-molecule measurements, we connect what we observe to the underlying physical forces, turning observations into mechanisms. This interplay between experiment and computation is central to how our lab uncovers the rules governing membrane proteins.
Light and oxygen are essential to life, yet together they can quietly damage the proteins that keep our cells running — a process linked to aging and to disorders of light-exposed tissues. How this damage unfolds at the molecular level, and why some hidden parts of a protein are vulnerable, is still not well understood. By bringing single-molecule tools together with a range of other techniques, we study how proteins are attacked from within, uncovering damage routes that the conventional view overlooks. This opens a new way of thinking about how proteins age and fail.
