Interactive Curriculum Vitae for
| rch | eong@jh | u.ed | u |
NF-κB is a principal transcription factor that mediates inflammation, and its abnormal activity has been implicated in numerous diseases such as asthma and cancer. I combine mathematical models and experiments to understand how this important protein responds to external signals, such as tumor necrosis factor (TNF). In an initial study, I developed and used a detailed mathematical model, encompassing the biochemical reactions involved in TNF-induced NF-κB activity, to understand why in experiments NF-κB is observed to respond sensitively to a wide range of TNF doses. The model predicted that (1) transient activity of the signaling intermediate IKK was essential for the dose response, and this was subsequently verified experimentally, and (2) the observed dose response could be important physiologically by allowing TNF to robustly signal to cells across a large distance. (See J Biol Chem publication, below.)
My current studies are focused on studying TNF-NF-κB signaling in single cells. I use microfluidic devices to collect single cell signaling data in a high-throughput manner, and use computer models to analyze and interpret the resulting data. I anticipate that this systems biology-oriented approach will lead to novel and important insights into the function and regulation of NF-κB signaling, and may lead to new ways to diagnose and treat NF-κB-mediated diseases.
DNA, with its famous double helix, is a twisted structure. As such, all living organisms have mechanisms to modify its twisting, bending, and unwinding, and these mechanisms are important in the control of gene expression. I studied how DNA shape is altered by Lac repressor, a bacterial protein well-known to alter DNA shape upon binding, thereby inhibiting the expression of genes needed for lactose metabolism. Since Lac repressor is a homodimeric protein in which each dimer can bind to a specific DNA sequence, a single Lac repressor protein can bind to a single piece of DNA in two locations. This forces the intervening DNA to adopt a looped structure. I adapted existing Monte Carlo methods for modeling the shape of naked DNA in order to model DNA bound to Lac repressor. My computations suggested that there were only a few common shapes for the looped DNA, especially an "open" relaxed loop and a "closed" compact loop, depending on the DNA sequence and orientation of the protein dimers. The predicted shapes correlated well with DNA cyclization experiments used to measure overall DNA topology and FRET experiments used to measure loop size. This study helped to elucidate principles of how DNA loops form, which help us understand how such loops affect gene transcription, and may in the future aid in the rational design of DNA-protein complexes in nanotechnology applications.
pp32 is a tumor suppressor protein which is highly expressed in stem-like cells and other populations of self-renewing cells. In in vitro assays, pp32 is able to inhibit transformation induced by overexpression of the oncogenes ras and c-myc. For my high school research project, I investigated whether pp32 is able to inhibit transformation of mutated forms of myc. I examined mutations that are commonly found in human cancer, specifically mutations near phosphorylation sites and within the transactivation domain of myc. I found that pp32 was able to inhibit some myc mutants, in a dose-dependent manner, but not all of the mutants. This differential inhibition shows that suppression of myc by pp32 depends on the integrity of specific domains of myc, and may suggest specific mechanisms/pathways by which pp32 exerts its effects.
This article comprehensively reviews how computer models, integrated with experiments, have been used to great effect to understand the molecular mechanisms that regulate the activity of NF-κB, an important inflammatory transcription factor.
This article reviews how quantitative computer models, along with experiments, are an extremely useful tool in studying signaling pathways. In particular, we illustrate how models have yielded important insights about the NF-κB and MAPK pathways.
An epigenetic change (loss of imprinting, LOI) to the gene for insulin-like growth factor II (IGF-II) has been linked to an increased risk of developing colon cancer. This paper suggests a mechanism which may account for this observation: LOI is associated with increased sensitivity to IGF-II and increased expression of proliferation-related genes. Furthermore, a drug that specifically targets IGF-II signaling substantially reduces the formation of aberrant crypt foci in LOI mice, suggesting a new chemopreventive strategy for colon cancer. For this study, I helped measure IGF-II signaling sensitivity, using a novel microfluidic assay.
This paper is a case report of a patient who was under our care during my neurology clerkship. The patient was addicted to eating mothballs (paradichlorobenzene, PDB) and suffered from withdrawal due to hospital-enforced abstinence. We describe the syndrome of PDB withdrawal based on our patient and the few other reported cases of PDB addiction.
This paper describes an analysis of the TNF-NF-κB signaling pathway that iterates between computer models and experiment. Experimentally, we found that the temporal profile of NF-κB is identical across a wide range of TNF doses. A computational model, used to understand the molecular mechanisms allowing for this phenomenon, strongly predicted that this was a result of highly transient activity of the IKK signaling intermediate. Experiments semi-quantitatively confirmed this prediction. Further exploration with the model suggest that the pathway properties help ensure that TNF can be used to communication between cells across long distances in tissues.
This paper describes a computational analysis of DNA loops induced by Lac repressor. I used Monte Carlo methods to simulate the shape of short loops of DNA with sequence-directed bends. The simulations predicted two main shapes: a "closed" loop with positive writhe and an "open" loop with almost no writhe. The predicted shapes correlate well with results obtained by DNA cyclization and FRET experiments. The simulations further predict that DNA-bound Lac repressor strongly prefers an open conformation, in which its dimers point away from each other, rather than the closed conformation suggested by crystallography, in which the dimers form a sharp "V".
This is a peer-reviewed technical comment that appeared in Science. We argue that sustained oscillations in TNF-induced NF-κB activity observed in live cell experiments (Science, 306:704) are an artifact of poorly controlled overexpression of NF-κB or its inhibitor IκBα. We show that (1) computer models predict that overexpression artificially enhances oscillations, (2) NF-κB activity measured in single wildtype cells is not consistent with sustained oscillations, and (3) sustained oscillations do not have functional consequences on the timing of gene expression.
This paper describes fluorescence energy resonance transfer (FRET) experiments used to measure the distance between two Lac repressor operators in a DNA loop. This revealed two types of loops: an "open" configuration in which the operators were far apart, and a "closed" configuration in which the operators were close together. The results correlated well with Monte Carlo simulations of loop shape.
This book chapter is a comprehensive review of the NF-κB transcription factor.
The MSTP is awarded to most MD-PhD candidates accepted to Johns Hopkins University and it provides support for MD-PhD training.
The Othmer National Scholarship is awarded to only 15 students each year, recognizing academic achievement and contributions to AIChE student chapters. I served as the treasurer (2001-2002) and secretary (2000-2001) of the University of Maryland AIChE student chapter.
The HHMI fellowship provided a research stipend and funds to defray research expenses, and was awarded by the University of Maryland on a competitive basis.
The Banneker/Key scholarship is the highest merit-based scholarship awarded by the University of Maryland and it fully covers tuition, room, board, and books.
Biotech YES is a competition that promotes entrepreneurship among graduate students and postdoctoral scientists. Teams that participate learn about the basics of how biotechnology ideas are commercialized and are judged based on the ability to formulate and present a business plan for a fictitious business. In 2004, the British Council sponsored teams from the USA and Canada. My team (which included David Noren, Saurabh Paliwal, and Blanka Sharma, all from Johns Hopkins) won among all North American participants, for a business that would commercialize a handheld retinal camera that would allow physicians to more easily inspect patients' eyes.
This is the highest honor awarded by the School of Engineering at the University of Maryland. It recognizes academic excellence and leadership. I served as President (2001-2002) of the Engineering Student Council, and in this capacity I served as a student liaison to the Dean and oversaw numerous large School-wide student events.
The IOCCC is one of the longest-running programming contests held over the Internet (since 1984). The contest recognizes functional but creatively obscured C code, and winning entries often display unusual ingenuity and/or technical mastery of the C language. My entry in 2000 used the preprocessor to disguise a prime number generator in pseudocode, and won for "Best Abuse of the C Preprocessor". My 2001 entry, less than 400 bytes long, computed square roots to arbitrary precision and won for "Best Short Program".
The Putnam Exam is a premier mathematics competition for college students.
The award recognizes "outstanding academic, artistic, or leadership endeavors" in addition to academic excellence. Second team honors are given to the top 40 students in the nation.
The ISEF is a premier science fair competition for high school students. I qualified for the ISEF by winning the biological sciences category of the Baltimore Science Fair. My project was titled "Characterization of pp32-myc interaction".
The Science Talent Search (now sponsored by Intel) is a premier competition recognizing excellence in science research by high school students. My project was titled "Phosphoprotein 32 (pp32) variably inhibits lymphoma-derived myc mutants".
I initiated a three-year project to jump-start after-school math activities for high schools in Baltimore County, MD. The two main activities are the Baltimore County Math League (BCML) and the Baltimore County ARML Team. The BCML consists of math clubs at participating schools that practice problem-solving on a weekly basis, and compete against each other in monthly contests. The ARML team consists of the most talented students from throughout the county, especially those identified through BCML contests, and the team competes at the prestigious national contest, the American Regions Math League (ARML). I work closely with the Office of Mathematics of the Baltimore County Public Schools, teachers throughout the county, and many volunteers to implement the activities.
The development of the BCML and ARML team has proceeded steadily with great success. In the first year, we piloted the ARML team. We recruited students from 8 different schools, held practices for two months, scrimmaged against the Howard County ARML Team, and sent a full team (15 students) to ARML. In the second year, we piloted the BCML in four schools and held four league contests. We again recruited for ARML and sent (in conjunction with the Maryland Area Homeschoolers) two teams to ARML. Now in the third year (2007-2008), the BCML has expanded to 6 schools and we are again preparing to send two teams to ARML. Additionally, we are transitioning so that the Office of Mathematics takes full control of the league and ARML team in the future, thus achieving the original goal of making these activities permanent.
Some notable accomplishments:
I created a central website to collect resources useful in the basic science, clinical science, and clinical years of medical school. The website is used heavily by students at Johns Hopkins and elsewhere, and receives hundreds of page hits per day. Additionally, the website, whose content is completely due to student contributions, helps to maintain a highly collegial and cooperative atmosphere within the Hopkins medical school.
I devised an algorithm, grounded in first principles, to rank men's college basketball teams. The rankings consistently predict the winner in 72-73% of all games. Performance is comparable to Vegas and brackets based on the rankings consistently place in the 90th percentile or above in the ESPN Tournament Challenge. The methodology can be applied to many other sports, including football and baseball.