Meet the 2025 Graduates of the Gerstner Sloan Kettering Graduate School

I conducted my dissertation research in the laboratory of John Maciejowski, PhD, whose work centers on cytosolic DNA sensing and innate immune signaling in chromosomally unstable cancers. These cancers frequently exhibit mitotic errors, leading to the formation of micronuclei — small, aberrant nuclear structures that encapsulate mis-segregated chromosomes or chromosomal fragments. Due to their fragile nuclear envelopes, micronuclei are prone to spontaneous rupture, exposing DNA to the cytosol. This aberrant DNA is sensed by the innate immune system, activating cytosolic DNA sensing pathways and triggering type I interferon responses that can alert and engage the immune system. Harnessing this innate immune activation presents a promising therapeutic avenue to enhance tumor immunogenicity and sensitize tumors to immune checkpoint blockade.

Cancers evade immune surveillance in part by degrading cytosolic DNA through the exonuclease TREX1. Previous work from our lab demonstrated that TREX1 localizes to ruptured micronuclei, where it degrades micronuclear DNA and suppresses cGAS-STING–mediated innate immune signaling. This localization relies on TREX1’s tethering to the endoplasmic reticulum (ER); however, the mechanism by which ER tethering enables TREX1’s recruitment and catalytic activity is unclear. My thesis research aimed to elucidate the molecular basis of this ER-dependent localization and function of TREX1.

To uncover factors critical for ER-tethering–dependent TREX1 localization, we investigated proteins enriched at ruptured micronuclei. Using proteomic analysis of purified ruptured micronuclei, I identified a barrier to autointegration factor (BAF) and a class of LEM domain proteins known to interact with BAF. BAF is a small DNA-binding protein that plays a key role in nuclear envelope repair through its interaction with LEM domain proteins. Using live-cell imaging with fluorescently tagged BAF and inducible BAF knockout cells, I found that BAF is highly enriched at ruptured micronuclei, and its presence is essential for the recruitment of ER membranes and the localization of TREX1. Further analysis revealed that BAF mediates this process through its interaction with LEM domain proteins, thereby facilitating the access of ER-tethered TREX1 to micronuclear DNA substrates. These findings reveal the molecular pathway by which TREX1 is targeted to ruptured micronuclei.

Interestingly, while BAF facilitates TREX1 recruitment to ruptured micronuclei, it also plays a paradoxical role in restricting TREX1-mediated DNA degradation. I found that BAF binding to micronuclear DNA inhibits TREX1 resection, highlighting BAF’s protective function. Upon BAF depletion, micronuclear DNA exhibited enhanced resection, underscoring its role in shielding DNA from nuclease activity. Although this seems contradictory — BAF both promotes TREX1 localization and blocks its activity — it can be explained by the context. 

In primary nuclei, where rupture events are typically transient, BAF binds to chromatin and helps protect genomic DNA from damage. In contrast, ruptured micronuclei often remain chronically exposed, and BAF’s continued presence inadvertently enables TREX1 access via interactions with LEM domain proteins and TREX1 ER tethering. Supporting this model, a TREX1 mutant lacking ER tethering restores resection activity in the absence of BAF, whereas resection is fully blocked when TREX1 ER-tethering mutant is in the presence of BAF. These findings reveal that ER tethering allows TREX1 to bypass the BAF-mediated barrier and access micronuclear DNA.

Finally, I investigated how BAF and TREX1 influence cGAS-STING–mediated innate immune signaling. Depletion of BAF in chromosomally unstable cancer cells led to elevated cGAS activation, while simultaneous depletion of both BAF and TREX1 caused a dramatic increase in cGAS activity. These results identify BAF and TREX1 as negative regulators of the cGAS-STING pathway and type I interferon signaling. Their presence helps explain why irradiation-induced micronuclei often require prolonged mitotic progression — sometimes over a week — to elicit a detectable interferon response, a previously unresolved question in the field. These findings also present a promising therapeutic opportunity: targeting negative regulators like BAF and TREX1 could amplify innate immune activation, boost anti-tumor immunity, and enhance the efficacy of immunotherapies.

My PhD training in the Maciejowski lab and at GSK has provided me with a strong foundation in scientific research, combining rigorous academic inquiry with translational insight. I am excited to bring this experience into the next stage of my career, where I aim to contribute to drug discovery efforts in the biotechnology industry.

I conducted my PhD work in the laboratory of Morgan Huse, PhD, an interdisciplinary group that spans the fields of synthetic chemistry, materials science, biophysics, cell biology, cancer biology, and immunology. During my time there, I developed an exciting new technique to study how T cells kill their “targets” — infected cells and cancer cells. 

First, I learned how to synthesize small hydrogel microspheres that are soft enough for individual immune cells to deform (think “stress balls” for cells). I then modified this method to visualize with microscopy what killer T cells do to the surfaces of their targets. My method showed for the first time that T cells exert two categorically different types of physical forces on their targets. First, T cells compress them — that is, they squish and dig into them. At the same time, they micropattern the target cell membranes through beautiful, wavelike motions of the T cell cytoskeleton. Surprisingly, these mechanical forces are important for successful killing, and weakening either type of force also blunts a T cell’s ability to kill cancer cells. Our work was published in Science Immunology in 2024. 

During this journey, I learned several important lessons well beyond all the technical ones: 

My GSK chapter has been incredibly rewarding and meaningful; it’s really second to none in terms of giving us the tools to grow and succeed as scientists. These days, I’m a postdoctoral researcher at Vanderbilt University in Nashville, building new kinds of microscopes to drive unique experiments in bioimaging.

My research in the lab of Ross Levine, MD, looked at how inflammation — our body’s natural response to injury or infection — can influence changes in blood stem cells as we age. Sometimes, these stem cells develop mutations and start multiplying more than they should, a condition known as clonal hematopoiesis (CH). CH becomes more common with age and can increase the risk of serious diseases, including cancer and heart disease, among other age-related conditions. To understand this better, I used genetic tools and bred mice to study how inflammation might encourage CH cells to grow — and how that growth might eventually lead to disease. 

I also studied the immune systems of older mice with these mutations to find ways we might be able to stop this process before it causes harm during aging. My findings could help scientists develop new treatments to prevent age-related diseases and improve health as people get older, when mutant clones are detected in their blood. With more and more people living longer, this research could play an important role in promoting healthy aging and reducing the risk of age-related conditions as well as intercepting progression to cancer. 

One key lesson I’m taking with me is that while perfection may not be attainable, striving for excellence always is. Putting in your best effort, staying focused, taking care in your work, and taking care of yourself is what really matters. Growth comes from staying committed to the process, learning from mistakes, showing up with integrity, and learning when to say no politely while still being collaborative. That’s where the most meaningful progress — and satisfaction — happens: Keep going, doing your best, and being forgiving when intentional tradeoffs need to be made. 

I am completing a short postdoctoral stay at Cold Spring Harbor Laboratory as an NIH/NCI K00 fellow. I will be transitioning into healthcare/life sciences consulting this summer. 

During my PhD work in the lab of Ming Li, PhD, I dedicated my research to studying T helper (Th) cells within the peripheral immune system. These cells play a crucial role in orchestrating immune responses. Under normal conditions, their activity is tightly controlled by tolerance mechanisms to prevent harmful attacks on our own tissues. However, cancer cells exploit these tolerance pathways to evade immune detection and sustain unchecked growth. My work focused on investigating the role of two key immune pathways — TGF-beta and PD-1 — and how their interactions regulate Th cells in immune homeostasis and during liver cancer progression. 

I uncovered the underlying mechanisms that lead to systemic autoinflammation when both TGF-beta and PD-1 signaling pathways are disrupted in Th cells. Using genetically modified mouse models, I leveraged high-dimensional flow cytometry and single-cell transcriptomics to dissect the cellular interactions driving immune dysregulation. A crucial finding from this research was identifying the CD40-CD40L interaction between professional antigen-presenting cells and Th cells, which sustains a positive feedback loop that exacerbates the immune response. Additionally, we pinpointed tissue macrophages as the critical source of PD-L1, the ligand for the PD-1 receptor on T cells. These discoveries were essential in understanding how Th cells contribute to autoimmunity and cancer progression. 

Building on these findings, I conducted proof-of-concept studies that explored the therapeutic potential of co-targeting the TGF-beta and PD-1 pathways. Using an in-house designed bispecific antibody that combines PD-L1 inhibition with TGF-beta signaling disruption specifically on Th cells, I demonstrated significant therapeutic benefits, particularly in unleashing immune responses capable of targeting cancer cells. This combination therapy showed promising effects in restoring immune function and limiting tumor growth, offering new insights into immune modulation for cancer treatment. 

After completing my PhD, I joined Genentech as a scientist in the translational medicine department. I apply my extensive knowledge of immunology and computational analysis to leverage clinical data for discovering and validating new therapeutic targets, particularly in the context of inflammation-related diseases. I am excited to continue expanding my expertise in translational research and contribute to the development of innovative therapies that bridge the gap between basic science and clinical applications, advancing patient care and treatment outcomes. 

I conducted my thesis research in the lab of Richard White, MD, PhD, which studies melanoma through the lens of cellular metabolism and the tumor microenvironment. My research focused on the role of lipid metabolism in melanoma development and progression. 

When I first joined the lab, they had recently published a paper demonstrating that during tumor progression, melanoma cells take up free fatty acids released by adipocytes (fat cells) in the tumor microenvironment (TME) and store them in specialized organelles called lipid droplets. This led us to ask: What is the role of lipid droplets in melanoma? 

To better study the role of lipid droplets in the tumor and TME, we built transgenic zebrafish and melanoma cell lines that express a fluorescent lipid droplet reporter. This approach allowed us to image lipid droplet dynamics easily in live animals — across a variety of cell types — and in cell culture. It will also facilitate future work investigating the regulators of lipid droplets in melanoma. 

Another way to understand the role of lipid droplets in melanoma is to study their components and how these components might be regulated. Lipid droplets are composed primarily of a lipid-rich core; however, this core is surrounded by a membrane studded with proteins. These proteins are known to regulate diverse aspects of lipid droplet formation and function. We profiled this protein envelope and discovered that specific proteins were enriched at the lipid droplet in particular melanoma cell states. Notably, high levels of one protein, DHRS3, were sufficient to push cells toward a more undifferentiated/invasive state. Our data demonstrated that the dynamic composition of the lipid droplet protein envelope can regulate melanoma cell state. 

Since defending my thesis, I have been working as a senior associate consultant at Lumanity. I solve strategic challenges facing biotech and pharmaceutical companies in the life sciences. The scientific expertise, critical thinking, and problem-solving skills I developed during my PhD continue to serve me well in this role, and I am incredibly grateful for my time as a GSK student. 

My thesis research in the labs of Richard White, MD, PhD, and Michael Overholtzer, PhD, focused on melanoma, an aggressive skin cancer driven by genetic and environmental factors. While advances in genetic sequencing have identified thousands of genes that may contribute to melanoma, validating their functions in living organisms remains a challenge. My PhD research aimed to develop new approaches to study the complex genetic mechanisms underlying melanocyte and melanoma biology using zebrafish as a model organism. 

My dissertation began with investigating the role of CLMP, an immunoglobulin cell-adhesion molecule, in melanoma initiation and progression. We found that CLMP has diverse functions, promoting melanoma growth through cell-intrinsic and extrinsic mechanisms, yet acting as a tumor suppressor in other cancer types. 

To better understand the effects of CLMP loss in melanoma, we needed a lineage-restricted knockout approach, as a traditional global knockout led to unintended tumors in reproductive and gastrointestinal tissues. To address this challenge, we developed a highly efficient method for melanocyte and melanoma-specific knockout in zebrafish by inserting Cas9 at the mitfa locus, a master regulator of melanocyte development. 

Using this system, we identified new roles for pigmentation-associated genes from the UK Biobank and developed a platform to screen for tumor-promoting and tumor-suppressing gene functions in melanoma. Importantly, we distinguished direct (cell-autonomous) versus indirect (non-cell–autonomous) gene functions, revealing the complex cellular networks that exist in normal and disease states. 

Throughout my PhD, I had the privilege of collaborating with brilliant scientists across disciplines and receiving mentorship from researchers whose guidance was instrumental to my success. My experiences at GSK shaped my passion for advancing biomedical research. After defending my thesis, I joined the Pershing Square Foundation as a life sciences program manager, where I now help fund cutting-edge research in cancer and neurodegenerative diseases, supporting innovative solutions to some of the most pressing medical challenges.

I conducted my thesis research in the laboratory of Ross Levine, MD, a member of the Human Oncology and Pathogenesis Program and Senior Vice President of Translational Research at Memorial Hospital. Research in the Levine laboratory aims to identify and characterize somatic mutations that drive myeloid malignancies, with a focus on the role of signal transduction and epigenetic cooperativity in leukemic transformation.

Mutations in genes that regulate signaling and epigenetic pathways commonly co-occur in myeloid malignancies. The co-mutation of IDH2 and NRAS represents a clinically relevant context to study cooperative pathogenicity between these mutation classes. My first project was aimed at better understanding how IDH2 and NRAS mutations promote pre-leukemic outgrowth. Using genetically engineered mouse models, we found that pre-leukemic cells with mutations in Idh2R140Q and NrasG12D were highly proliferative, competitively fit, and possessed/promoted inflammatory features. This work provided new insights into the molecular underpinnings of Idh2R140Q- and NrasG12D-driven hematologic disease.

Progressive transformation to acute myeloid leukemia (AML) is associated with dismal outcomes. While many AML patients receiving intensive chemotherapy achieve a remission, most individuals relapse due to the presence of measurable residual disease (MRD). Reliable detection of MRD in AML is a major obstacle in achieving durable cures, as current MRD testing methods face limitations in assay sensitivity, technical precision, and broad applicability. Developing assays that are sensitive, specific, and conducive to studying MRD biology is necessary to guide the development of more effective curative treatments. 

My second project was focused on developing a single-cell MRD (scMRD) assay by combining flow cytometric enrichment of targeted hematopoietic precursors/blasts with integrated single-cell DNA sequencing and immunophenotyping. Our scMRD assay enabled sensitive MRD detection and illuminated the clonal architecture of leukemic cells surviving AML therapy. Our study lays the groundwork for deciphering the clonal characteristics of MRD in AML.

After my PhD, I will be joining Mizuho as a biotech equity research associate.

My PhD thesis work in the lab of Morgan Huse, PhD, focused on the mechanical interactions between effector immune cells and their targets, with a specific focus on macrophages. Macrophages are versatile immune cells that play multiple roles in human biology. Their main function, called phagocytosis, involves binding, engulfing, and digesting various particulate matter, including cellular debris, live pathogens, and tumor cells. Each of these targets has very different physical properties, and it had been previously observed that macrophages are “mechanosensitive,” meaning that phagocytosis is more likely to occur on stiff particles and less likely to occur on soft particles. We wanted to understand how and why this occurs because cancer cells tend to be quite soft. This may allow them to evade macrophage-based immune responses.  

To do this, we optimized an experimental system using hydrogel microparticles that can be tuned to different stiffnesses and measured how macrophages respond to them. We found that a class of cell-adhesion receptors called Beta2 integrins are critical for mechanosensing, and that they create a mechanical “checkpoint” that prevents phagocytosis of soft particles. This finding was especially exciting because it may enable scientists to engineer mechanically insensitive macrophages that can engulf and clear cancer cells despite their softness.  

I will always be grateful for GSK’s unique program structure and commitment to bringing in students from diverse scientific backgrounds. In my PhD experience, I had the opportunity to work at the interface of cell biology, immunology, materials science, and bioengineering. This was enabled by the collaborative and interdisciplinary community that GSK has fostered. 

Since the completion of my PhD, I have been working as a postdoctoral fellow at Regeneron Pharmaceuticals within the genetic medicines division, where I am working to uncover molecular mechanisms that enable viral vector gene therapies.  

My research dissertation, done in the lab of Philipp Niethammer, PhD, was broadly focused on mechanical biology, a new and exciting area of research where physics and biology intersect. Before I joined Dr. Niethammer’s lab, the team had discovered a type of phospholipase protein called cytosolic phospholipase A2 (cPLA2), which binds to the nuclear membrane and produces inflammatory lipid mediators when the membrane is physically stretched. (This stretching is due to disruption of osmotic balance in the zebrafish epithelium after injury.) But at that time, nobody understood how a physical change in the membrane could lead to a shift in localization of a nucleoplasmic protein. 

This is where my research came in to fill the gap. The first aim of my research was to specifically understand the biochemical and biophysical mechanism of cPLA2 and related peripheral proteins that sense membrane stretching. Using an interdisciplinary set of research techniques — ranging from artificial liposomes and fluorescently labeled cell lines to zebrafish models — my research revealed that hydrophobic and electrostatic residues on cPLA2 synergistically facilitate the protein’s ability to sense membrane tension. Stretching the nuclear membranes increases the spacing between neighboring lipids, allowing the hydrophobic residues of cPLA2 to access and insert into the membrane. At the same time, stretching enhances cPLA2 calcium sensitivity, allowing the protein to bind to nuclear membranes under stringent calcium conditions. 

I also discovered that the capability to sense membrane tension is an overlooked feature of peripheral proteins, as many other proteins have shown behavior similar to cPLA2 — although to a lesser extent. Overall, my research was the first of its kind to show that peripheral proteins can sense physical tension. This finding not only offers new insights into how physical forces influence biology, but it also opens the door to developing noninvasive membrane tension sensors in the future. 

The second aim of my research focused on the role of ER membrane in the rise and maintenance of nuclear membrane tension. It remains a mystery in cell biology how nuclear membrane tension arises, rather than being dissipated by surrounding continuous ER membrane. Therefore, I wanted to address this enigma. Surprisingly, I found ER membranes are fragmented and dislodged from nuclear membranes when exposed to physical force. This significantly reduces the membrane areas under stretch and allows nuclear membrane tension to build up. Overall, this research revealed a hidden function of ER membranes in regulating the mechanosensitivity of nuclear membranes. 

Deciding to complete my graduate training at GSK is the best decision I’ve made so far. My training at GSK combined the best of both worlds: learning about cancer progression and treatment from top clinicians and gaining a deep understanding of how biology works from leading scientists. The things I gained from my training — scientific rigor, teamwork, and fearlessness in exploring bold ideas — gave me the confidence and foundation for my current role as a postdoctoral fellow at Johns Hopkins, where I study the crosstalk between immune cell metabolism and the physical properties of the cancer microenvironment.

The lab of Jayanta Chaudhuri, PhD, studies B cells across all contexts. I joined the lab in 2017, eventually publishing my thesis on B cell leukemic transdifferentiation. 

B cells are responsible for generating specific antibodies that protect us against specific infectious agents (including viruses and bacteria). Their development progresses through a sequence of well-characterized stages, ultimately producing mature B cells that each possess a unique B cell receptor (BCR). Developmental rearrangement is the molecular mechanism that underlies the formation of the BCR. During that process, these cells divide many times and can transform into one of the precursor-B cell subtypes of acute lymphoblastic leukemia (ALL), the most common childhood cancer. 

Under the right conditions, this tumor can mutate into alternative lineages, called transdifferentiation, particularly to T cell or myeloid leukemias. Knowing the mediators of the rare event would allow us to predict, and potentially prevent, this mechanism of escape from therapies targeting the cancer. We developed a mouse model of this conversion process and characterized the resulting T cell leukemias, identifying several promising therapeutic avenues for future research. 

Building on the B cell biology and coding tools I learned as a student, I will spend a year setting up spatial transcriptomics (sequencing of cells in their localized context) in the Chaudhuri lab. In collaboration with a surgeon connected with the lab, we will generate a pipeline to study and analyze B cells in breast cancer. 

I completed my dissertation research in the laboratory of Ross Levine, MD, which focuses on the development of new treatments for serious blood cancers. My research is focused on identifying potential therapeutic targets in acute myeloid leukemia (AML). 

AML is frequently preceded by a recently discovered premalignant state known as clonal hematopoiesis (CH). Mutations in chromatin modifier genes have been shown to occur in hematopoietic stem cells (HSCs), resulting in an aberrant expansion of these mutant cells and ultimately an increased risk of developing blood cancers. Despite extensive research documenting the impact of these mutations, few targeted therapies exist for the prevention or treatment of these diseases. 

For my dissertation project, I performed CRISPR screens on HSCs harboring mutations frequently found in CH and AML. Since large-scale screens on HSCs have been historically unfeasible, I developed a co-culture system of HSCs with bone marrow endothelial cells to enable these studies. Using this system, I identified KDM3B and its downstream pathway as promising new therapeutic targets in a genetically defined subset of CH and AML.

Beyond this specific target, my work has broader implications in providing a path for new target discovery in mutant HSCs, which can inform the development of novel therapies for the prevention and treatment of blood cancers. 

GSK and Dr. Levine’s mentorship provided me with an excellent training environment in translational cancer research. After graduation, I will transition to working in biotech research and development as a scientist at Arena BioWorks. 

Embryonic development follows a sequence of events that is broadly conserved across species, yet the pace of development is highly variable and particularly slow in humans. Variations in developmental timing across species is thought to be integral to the determination of organism size, lifespan, and tissue complexity. Species-specific differences in developmental pace are largely recapitulated in stem cell models in vitro, suggesting a cell-intrinsic clock that tracks time. Understanding the molecular nature of intrinsic mechanism that controls timing is pivotal for understanding fundamental biological phenomena with broad implications in stem cell therapy and disease modeling. 

During my dissertation research in the lab of Lorenz Studer, MD, I focused on identifying genetic factors that regulate the timing of the human differentiation process. Using the directed differentiation of human embryonic stem cells (hESCs) into neuroectoderm as a proxy for developmental timing, I performed a whole-genome CRISPR-Cas9 knockout screen and found that the epigenetic factors menin and SUZ12 modulate the speed of neural differentiation. Genetic and pharmacological loss-of-function of menin or SUZ12 accelerate the acquisition of neural fate by altering the balance of activating H3K4me3 and repressive H3K27me3 chromatin modifications at bivalent promoters of developmental genes — thereby priming them for a faster activation upon differentiation. 

Our study further revealed a synergistic interaction of menin and SUZ12 in modulating the speed of developmental programs. The acceleration effects were similarly observed in definitive endoderm, cardiomyocyte, and neuronal differentiation paradigms, pointing to chromatin bivalency as a general driver of timing across germ layers and across developmental stages. 

GSK provides the opportunity to explore the full spectrum of biomedical research — from fundamental discoveries to clinical applications. During my PhD journey, the broad exposure to mechanistic investigation of biological phenomena and the development of innovations aimed at improving human health and aging has inspired me to pursue a career dedicated to developing innovative therapies for aging-related diseases.  

I conducted my thesis research in the lab of Danwei Huangfu, PhD, studying the effect of gene dosage. Studies of gene function usually rely on generating knockouts, where the function or expression of a gene is completely disabled. However, in physiological contexts, a gene’s expression is not always all-or-none. Indeed, the vast majority of disease-associated genetic variants affect non-coding regions, such as enhancers, that fine-tune gene expression, rather than protein-coding regions. Therefore, there is a pressing need to understand the implications of variations in gene expression levels on human health and disease. 

To address this gap in knowledge, I used human embryonic stem cells (hESCs) as a model to study gene dosage. These cells are hallmarked by their ability to balance exquisitely between self-renewal and differentiation and could therefore be particularly sensitive to changes in the expression levels of genes regulating these processes. The transcription factors NANOG and OCT4 are core regulators of hESC identity. They allow hESCs to self-renew and to maintain full differentiation potency, but they need to be down-regulated for the cells to differentiate. Therefore, decreasing the expression of NANOG and OCT4 in hESCs could both prime them for differentiation and reduce their ability to self-renew as fully potent cells, raising the question how hESCs fine-tune their responses to varying levels of these core transcription factors to ensure precise control of self-renewal and differentiation. 

I examined the impacts of different dosages of NANOG and OCT4 on hESC self-renewal and differentiation in two projects. In the first project, I mapped functional enhancers in hESCs and characterized the roles of two novel NANOG enhancers on regulating NANOG dosage. Enhancers are a crucial component of gene regulation, but systematic mapping of functional enhancers in hESCs was lacking. To address this gap in knowledge, our lab performed a CRISPR screen targeting putative enhancers of NANOG and OCT4, among other transcription factors. This screen uncovered hits including known enhancers of OCT4 and two previously uncharacterized enhancers of NANOG. To study these novel enhancers’ role in controlling NANOG dosage and ESC identity, I used CRISPR-Cas9 and paired gRNAs to delete these regions. However, I was only able to generate heterozygous and not homozygous deletions for these two enhancers, suggesting they are both essential to hESCs. Deleting a single copy of either enhancer decreased NANOG expression by 20% to 30%, compromised hESC self-renewal, and increased their propensity for differentiation. This work demonstrates that the NANOG enhancers, through maintaining the full dosage of NANOG expression, also maintain the balance between hESC self-renewal and differentiation. 

In the second project, I sought to systematically examine cellular responses to varying NANOG and OCT4 dosages. To achieve a broad and finely divided range of these genes’ expression levels, I employed a library of gRNAs targeting NANOG and OCT4 enhancers with varying efficiencies. I then used single-cell RNA-seq to profile transcriptomic changes in response to the titrated perturbations in the contexts of ESC self-renewal and definitive endoderm (DE) differentiation. At the ESC stage, NANOG or OCT4 perturbations affected hESC self-renewal in a dosage-dependent manner, resulting in a reduced number of cells and altered expression of ESC and differentiation genes. 

To quantitatively understand the dynamics of gene expression changes, I collaborated with computational experts to perform dosage modeling. This revealed that ESC and differentiation genes displayed strong switch-like behavior in response to the perturbations, suggesting that hESCs exist in a precarious balance between being primed for differentiation and maintaining ESC identity. In DE, reduced levels of NANOG and OCT4 led to accelerated differentiation, misexpression of genes associated with alternative lineages, and cell dropout. 

Further modeling revealed that many DE-specific genes exhibit a non-monotonic response to decreasing dosages of NANOG and OCT4, being first up-regulated and then down- regulated. My discovery of the biphasic outcomes of differentiation, predicated on the dosage of NANOG and OCT4, suggests that this feature may be applicable across various stem cell types and can be harnessed to maximize differentiation efficiency. Together, my two projects underscore the sensitivity of cell states to the dosage of core regulator genes. They also provide a framework to study gene dosage effects underlying cell identity transformations not only from stem cell to differentiation, but also from health to disease. 

My PhD thesis research in the lab of Christine Iacobuzio-Donahue, MD, PhD, focused on genomic studies of pancreatic cancer, specifically pancreatic ductal adenocarcinoma (PDAC). My research spanned basic and translational aspects of pancreatic cancer, including projects such as studying the genomic underpinning of the disease and quantifying tumor features that correspond to therapy response or resistance. 

The primary output of my research was the development and application of a new single-cell DNA sequencing (scDNA-seq) system and associated computational analysis algorithms that better elucidated the disease’s genomic evolution. From a technical perspective, molecular characterization of pancreatic cancer has been hindered by low tumor cellularity as well as contamination from non-cancerous cells. The scDNA-seq method overcame both challenges.

We made several novel observations: 

• Different pancreatic cancers or different subclones within the same pancreatic cancer convergently evolve toward unresponsiveness to TGF-beta signaling.
• Continuous evolution manifested as focal deletion to tumor suppressor genes piece by piece or amplification of oncogenic genes through extrachromosomal-DNA (ecDNA)-mediated mechanisms.
• Pancreatic cancers in patients with germline BRCA2 mutations displayed a different path of genomic evolution than the canonical route. 

These observations furthered our knowledge of the pancreatic cancer’s genome in the absence of targeted therapy. 

In essence, we developed a powerful tool that enables higher-resolution genomic studies. This tool has now been applied to studying not only pancreatic cancer but also lung cancer, colon cancer, and neurodegenerative disease. This study established the ground-truth genomic evolution patterns of pancreatic cancer; it will form a crucial baseline for comparison as targeted therapies, spearheaded by KRAS inhibitors, are poised to sculpt different genomic evolution paths of this disease. 

My doctoral training in the Iacobuzio-Donahue lab brought me to the forefront of translational cancer research. After graduation, I plan to apply the knowledge and skills to innovative research and product development in the biotechnology industry. Through my time at GSK/MSK, I was fortunate to have connected with many brilliant scientists and clinicians who will be long-term friends and collaborators.