Current Fellows

Inwha Baek Jane Coffin Childs Fund Fellow

Laboratory of Mammalian Cell Biology and Development, The Rockefeller University

Sponsor: Elaine Fuchs

Profound alterations in gene expression profiles occur as stem cells from normal tissue transform into cancer stem cells (or tumor-initiating cells) that are able to initiate tumor growth and propagate tumor masses. However, how the gene expression program is rewired during tumorigenesis remains elusive. My research project centers on exploring and identifying key factors that are responsible for driving the divergence of their gene expression profiles. Using in vivo mouse models and genomics and imaging approaches, I am investigating how spatiotemporal genome organization plays a role in oncogenic transcriptional reprogramming during skin squamous cell carcinoma (SCC) development. Skin SCC has emerged as a public health issue due to its increasing incidence and potential for metastasis and recurrence, particularly for the patients placed on immunosuppressive drugs. If successful, my work would further our understanding of the mechanisms underlying oncogenic transcriptional reprogramming and provide new avenues for developing new cancer therapeutics.

Emily Bayer Jane Coffin Childs

Department of Biozentrum, University of Basel

Sponsor: Alex Schier

Intuitively, humans seem aware of the fact that visceral sensations are related to their emotional or stress perceptions of the world. English idioms such as the heart “leaping” reflect excitement, the heart “sinking” reflects despair, and “gut feelings” reflect intuition. This conscious awareness of the reactions of the viscera suggests a two-way relationship between perception of the body and reaction of the brain, but the biological underpinnings and relevant neural circuits are still understudied.

The zebrafish has the great advantage of external fertilization and optical accessibility during development, which I will utilize to study how sensory inputs from the body shape brain state and activity. I hope to understand the kinds of sensory information transmitted from the body to the brain, and the ways that this affects how the brain generates behaviors.

Tessa Bertozzi HHMI Fellow

Weissman Lab, Whitehead Institute of Biomedical Research

Sponsor: Jonathan Weissman

Chemical modifications to DNA and histones are implicated in the establishment of heritable cell type-specific transcriptional networks. The emergence of molecular epigenetic editors creates new opportunities to mechanistically probe these relationships and understand the functional repercussions of epigenetic dysregulation in cancer and aging. The CRISPRoff editor was developed in a collaboration between the Weissman and Gilbert labs as a single fusion protein containing the catalytically inactive dCas9, a repressive KRAB domain, and DNA methyltransferase domains. Transient expression of RNA-guided CRISPRoff achieves robust and heritable gene silencing in human cells, likely as a product of the synergistic spatiotemporal relationship between the coupled domains. Using a CRISPR-based screening approach, I plan to uncover and mechanistically characterize additional cooperative protein interactions which facilitate the establishment and maintenance of long-term transcriptional memory.

Amir Bitran Jane Coffin Childs Fund Fellow

Department of Molecular and Cellular Biology, University of California, Berkeley

Sponsor: Susan Marqusee

The majority of functions we associate with living thing are made possible thanks to the molecular functions of proteins. Proteins, just like cars and other macroscopic machines, require a specific 3D structure in order to be able to perform their functions and improper protein folding is linked to diseases cut as Alzheimer’s, Parkinson’s, and various cancers.  Yet despite decades of research, we still do not understand how proteins fold up into these structures, and what determines whether folding ultimately proceeds correctly—these questions are not addressed by structure-prediction algorithms such as DeepMind’s AlphaFold. It is crucial that we make progress on these issues if we are to rationally design treatments for misfolding diseases, and to predict evolution of organisms, which is often mediated by changes to protein folding and function.

All proteins are made up of one or more chains of amino acids. For some proteins, the physical and chemical interactions between these amino acids are entirely sufficient to drive protein folding into correct native structure.  But growing evidence suggests that, for many other proteins, these interactions instead cause the amino acid chain to misfold into non-functional molecular structures.  A major goal of my research is to understand how this conundrum is resolved in the complex cellular environment.

One possible resolution to this issue may lie in the fact that, in addition to folding, a protein molecule needs to be synthesized one amino acid at a time by the ribosome. It turns out that many proteins can start folding as they are being synthesized, a process known as co-translational folding which has been shown to significantly increase the odds that certain proteins fold correctly. Indeed, many proteins contain evolutionarily conserved slowdowns in their rate of synthesis at chain lengths corresponding to putative co-translational folding intermediates, indicating it is broadly useful to modulate synthesis rates to give time for co-translational folding. This is akin to how dance (analogous to a chain’s folding) is closely linked to musical rhythm (how quickly amino acids are added)—I may have taken this analogy a bit too far and written a musical piece inspired by it (The Dance of the Nascent Chain).

My research aims to develop a detailed molecular picture of this process, and why it is beneficial to fold co-translationally for many proteins, by combining in vitro and in vivo experimental techniques, physics theory and atomistic simulations. In the future, this interdisciplinary pipeline can also be applied to investigate additional complex processes in the cell including mechanisms of misfolding in disease.

John Blair Jane Coffin Childs Fund

New York Genome Center, New York University

Sponsor: Satija Rahul

Protein phosphorylation is a fundamental, dynamic process that can have drastic effects on cellular physiology. Mutations in kinases, the enzymes that phosphorylate other proteins, are often implicated in neurological disease. Understanding the context and consequences of protein phosphorylation in different cell types throughout neurodevelopment is imperative to developing new treatments as well as our basic understanding of cell biology. Recent technological developments permit the simultaneous quantification of protein levels, chromatin accessibility and gene expression from single cells (DOGMA-Seq). I am extending this technology to quantify both phosphorylated proteins and total proteins as well as chromatin accessibility and gene expression. I am applying this assay at discrete timepoints throughout in vitro neurodevelopment to reveal previously uncharacterized cell-type specific signaling patterns affecting gene expression and ultimately, cell fate decisions.

Paula Bucko Jane Coffin Childs Fund

Department of Systems Biology, Harvard Medical School

Sponsor: Galit Lahav

In response to DNA damage, the tumor suppressor protein p53 induces expression of stress-responsive genes to inhibit proliferation of cells with damaged DNA. Changes in p53 protein levels over time (p53 dynamics) impact cellular outcomes: p53 oscillations facilitate repair of DNA-damaged cells, whereas sustained levels of p53 promote senescence and cell death. While it is now established that p53 dynamics contribute to these competing cell-autonomous processes, how p53 dynamics regulate genes involved in non-cell-autonomous events, such as those involved in immune signaling, is not known. I propose to develop new tools and approaches to study the role of p53 in regulating immune gene expression in cancer cells and in mediating the killing of cancer cells by immune cells. This research will provide fundamental insights into the mechanisms that govern of cancer cell-immune cell interactions and pave the way for developing effective combination therapies to treat cancer

Vikram Chandra Jane Coffin Childs Fund Fellow

Department of Organismic and Evolutionary Biology, Harvard University

Sponsor: Mansi Srivistava

How animal brains evolved the capacity for sophisticated computation is not well understood. One major facet of this problem is the evolution of chemosensation. Chemosensation is the primary sense of most animals, and involves complex neural computations. We do not know how this sense evolved, or how most animals – which are aquatic invertebrates – perform chemosensation. I am studying chemosensation in an acoel worm, an aquatic invertebrate that by virtue of its phylogenetic position as the likely outgroup to all other animals with central nervous systems, retains some primitive features of early central nervous systems. Acoels nonetheless perform sophisticated behavior that requires complex chemosensory processing, but how their brains and chemosensors work is unknown. Using a combination of automated behavioral tracking, transgenics, and neural activity imaging, I aim to understand the logic of chemosensory processing in a tractable acoel worm. Through comparisons with known chemosensory mechanisms of other animals, this will shed light on how complex chemosensory systems evolved. This project will also establish experimental approaches for the future study of neural computations and behavior in acoel worms and other aquatic invertebrates.

Jingxun Chen Jane Coffin Childs

Department of Genetics, Stanford University

Sponsor: Anne Brunet

Reproductive aging is a global challenge. Older men and women face fertility loss and a higher chance of having children with genetic disorders. Currently, we lack a detailed molecular understanding of what causes reproductive aging in vertebrates. I am developing an emerging short-lived model system, the African killifish, to study vertebrate reproductive aging. The lifespan of this organism is 4 times shorter than mice and 7 times shorter than zebrafish. I will combine my graduate training (gamete biology) with the expertise of the Brunet Lab (killifish and aging) to probe the molecular basis of age-dependent fertility decline in the killifish and identify potential targets for therapeutic intervention. These studies will shed light on methods to protect or rejuvenate the germline from aging, which can have a profound impact on human fertility.

James Chen Jane Coffin Childs Fund Fellow

Department of Cell Biology, New York University, Grossman School of Medicine

Sponsor: Damian Eikiert - Co-Sponsor Darwin Hern

Antibiotic treatment of Mycobacterium tuberculosis (Mtb) is hindered by the inability of small molecules to cross the complex mycobacterial outer membrane (MOM). To understand the architecture of the MOM, I am using a combination of biophysical and microbiological approaches to study the MCE (Mammalian Cell Entry) proteins, which are critical virulence factors located on the Mtb cell envelope. These proteins are thought to transport hydrophobic molecules by assembling into large, multiplex structures that span the cell envelope and may have implications in MOM biogenesis and nutrient acquisition. However, the molecular bases for these functions are not known and the MCE proteins could play additional roles in the cell that have yet to be characterized. This work will provide mechanistic insights into these essential virulence factors that will be critical to understanding pathogenesis and to developing new antibiotics to counter emerging multi-drug resistant strains of Mtb.

David Colognori Jane Coffin Childs

Department of Molecular and Cell Biology, University of California, Berkeley

Sponsor: Jennifer Doudna

CRISPR-Cas systems provide prokaryotes with an adaptive immune mechanism whereby foreign nucleic acids are recorded and, when re-encountered, destroyed. Foreign DNA fragments are incorporated into the host’s CRISPR array and later transcribed and processed into crRNAs. crRNAs then assemble with Cas effector proteins and guide them to complementary nucleic acid sequences for destruction. The well-known Cas9 cleaves DNA site-specifically, and thus has been widely adopted as a programmable tool for gene editing. Analogous tools for cleaving RNA are lacking, with the exception of Cas13 which exhibits non-site-specific cleavage and toxic off-target effects. My research aims to discover and characterize new Cas effectors for precise RNA-cleavage in prokaryotes, and further develop them into tools for detection and cleavage of RNA sequences in eukaryotes.

Marlis Denk-Lobnig Jane Coffin Childs Fund Fellow

Department of Biophysics, University of Michigan

Sponsor: Kevin Wood

Cells and organisms such as bacteria or cancer cells live in communities and have complex population-level dynamics and emergent behaviors. These behaviors can significantly change their collective properties, including resistance to drugs and other environmental selective pressures. Predicting how cellular communities, rather than individuals, respond to drug exposure could help delay or prevent drug resistance during treatment.

Spatial structure and heterogeneity are important features of such communities – from the cell to the ecosystem scale.  In my research, I aim to understand the essential, but poorly understood, relationship between spatial structure and population-level dynamics in bacterial biofilm communities responding to antibiotic selection via theoretical and experimental approaches.

Frances Diehl HHMI Fellow

Department of Molecular Biology and Genetics, Johns Hopkins University

Sponsor: Laboratories of Rachel Green and Bin Wu

Protein synthesis is metabolically costly; it is therefore critical that cells regulate translation based on nutrient availability. Translation regulation must be flexible to enable cells to adapt to persistent nutrient deprivation, but still recover when nutrients are replenished. Signaling through the kinases GCN2 and mTORC1 can suppress both translation initiation and elongation in response to nutrient depletion, but how each mechanism contributes to global translational control is unclear. Here we investigate translation dynamics upon acute and long-term nutrient depletion, as well as during recovery from nutrient stress. In the immediate response to starvation, GCN2 robustly inhibits initiation to prevent ribosome loading onto transcripts. However, over longer periods of starvation, increased initiation causes translating ribosomes to collide with ribosomes stalled on transcripts. To explore these different temporal regimes, we employ a mass spectrometry-based approach to identify factors that modulate ribosome activity in response to nutrient stress through differential ribosome binding. This work will provide a more integrated understanding of how cells regulate translation and ribosome homeostasis across nutrient environments.

Caroline Doherty Jane Coffin Childs Fund Fellow

Department of OB/GYN, University of California, San Francisco

Sponsor: Diana Laird

Mammals with ovaries are born with a non-renewing supply of differentiated oocytes ranging from the thousands in mice to the millions in humans. While these high numbers imply a large stockpile, only a comparatively small number of the oocytes present at birth will ever be successfully ovulated and fertilized. To achieve this maturation, an oocyte must first be activated from its quiescent state and then undergo a period of extensive growth in order to accumulate large quantities of biosynthetic materials that are necessary to support the embryo prior to zygotic genome activation. We are currently limited in our understanding of factors that determine whether an oocyte will complete this growth or be fated for elimination/atresia. My research focuses on how both the intrinsic characteristics of oocytes and the extrinsic support provided by the surrounding somatic tissue determine whether an oocyte present at birth will ever be successfully ovulated. My goal is to apply this knowledge to future therapeutics for infertility.

Priscilla Erickson Jane Coffin Childs

Department of Biology, University of Virginia

Sponsor: Alan O. Bergland

Organisms exhibit diverse strategies to survive environments that vary in space and time. In temperate climates, environmental cues are used to anticipate the onset of unfavorable seasons. One of the most reliable indicators of season is photoperiod: the length of light and dark periods within a 24-hour day. Insects exhibit a spectacular array of responses to changes in season. For example, aphids develop sexually reproducing morphs in the fall, moths and lacewings develop unique seasonal patterns and colors, monarch butterflies undergo seasonal migrations, and hundreds of species, including the fruit fly Drosophila melanogaster, are able to suspend development or reproduction until more favorable conditions return. Despite nearly a century of research on insect seasonality and photoperiodism, the genetic pathways used to make these ecologically crucial transitions remain unknown. The abundance of genetic and genomic resources for Drosophila makes it an ideal study system for this question.

My research uses custom-built environmental chambers, field studies in an experimental orchard, and novel genetic mapping techniques to dissect the genetic basis of photoperiodism and seasonal responses in Drosophila. Understanding how insects detect photoperiod will inform our understanding of economically and biomedically important insects and offer predictions about how insects may adapt to ongoing anthropogenic climate warming in which temperature, but not photoperiod, is changing.

Abraham (Avi) Flamholz, Jr Jane Coffin Childs Fellow

Department of Biology and Biological Engineering, California Institute of Technology

Sponsor: Dianne Newman

While cells are often studied in suspension or monolayers, more structured forms like tissues and biofilms dominate natural environments. In such settings, the concentrations of critical nutrients like sugars and O2 vary in space and time because cells produce and consume them locally, leading to measurable differences in physiology and gene expression between nearby cells. Spatially structured environments therefore represent many-body systems interacting on multiple timescales through a rich collection of chemical and physical processes. My overriding goal is to determine whether metabolism in mixed biofilms can be predicted quantitatively from simple models with intelligible and measurable parameters. I am currently developing Pseudomonas aeruginosa, a model bacterium that grows in suspension and as a biofilm, as a model for studying metabolic heterogeneity in spatially structured environments. It is commonly assumed that variation in the local O2 concentration is a primary determinant of metabolic heterogeneity in biofilms. As such, I am developing optical approaches to measure local O2 concentrations in real time to test whether a mathematical model can explain O2 dynamics, cell growth, and metabolic rates in biofilms.

Nadezda Fursova Jane Coffin Childs Fund Fellow

Center for Cancer Research, National Cancer Institute/NIH

Sponsor: Daniel Larson

Enhancers are distal cis-regulatory elements that control precise execution of transcriptional programs during development and in response to external stimuli. How enhancers find and activate their target genes, and what molecular activities are required for enhancer function remains a central outstanding question in the field. Recent advances in nascent RNA-sequencing uncovered widespread transcription from enhancers, which has become widely recognized as a robust signature of enhancer activity. However, mechanistic understanding of enhancer transcription, its regulation and, most importantly, functional role in gene activation is currently missing.
In my work, I aim to address these fundamental questions by using single-molecule and live-cell imaging approaches to characterize the intrinsic dynamics of enhancer transcription in single cells. To generalize my conclusions from individual enhancers to a genome scale, my ultimate goal is to develop high-throughput single-molecule approaches for systematic characterization of enhancer transcription. Using these new tools, I will investigate how transcription at enhancers and their target gene promoters is coordinated at the single-cell level to discover if these processes are functionally linked. Together, this work will be an essential step towards a deeper mechanistic understanding of enhancer function in gene activation and how enhancer perturbations can lead to severe developmental disorders and cancer.

Annie Handler HHMI Fellow

Department of Neurobiology, Harvard Medical School

Sponsor: David Ginty

Our sense of touch emerges from a wide array of low-threshold mechanoreceptor neurons (LTMRs) that innervate the skin. Across our bodies, anatomically, molecularly and neurophysiologically distinct regions of skin house unique combinations of physiologically distinct LTMRs, which form unique 3-dimensional end organ structures as a result of their close association with local, terminal Schwann cells and other cell types. Distinct LTMRs and their end organ complexes, which are unique both in structure and mechanical sensitivity, allow us to perceive the slightest deflection of a hair follicle from a slight breeze, the low-frequency vibrations from the slip of your phone through your fingertips, or the high-frequency vibrations of a passing train. As such, our ability to perceive a wide array of mechanical stimuli in our environment emerges from the concerted activity of distinct end organ structures and the diverse LTMR endings embedded within. While the heterogeneity in both end organ structure and LTMR mechanical tuning has been appreciated for nearly 50 years, we understand very little of how the 3-dimensional structure of these end organs informs the tuning preference of their associated LTMRs and where the obligate mechanically sensitive ion channel, Piezo2, resides within them. As a postdoctoral fellow in the Ginty lab, I am using modern electron microscopy methods to create full-volume reconstructions of the distinct sensory end organs that enable our sense of touch. This 3-dimensional perspective, coupled with the genetic tools of the mouse and electrophysiological approaches, is allowing me to define molecular underpinnings of end organ function and to isolate the sites and mechanisms of mechanotransduction within the diverse sensory neurons of touch.”

Kiah Hardastle Jane Coffin Childs

Department of Organismal and Evolutionary Biology, Harvard University

Sponsor: Bence Olveczky

Understanding how the brain drives natural behavior is a central question in neuroscience. This quest is made particularly difficult by the fact that animal behavior is highly adaptable, thus requiring underlying neural circuits to alter the information they compute or represent depending on the task at hand. In my research, I examine how neurons in the motor pathway represent natural behaviors, and how these representations may change depending on the task the animal must perform. I investigate these questions using a combination of in vivo electrophysiology, machine vision, and computational models.

Patricia Horvath HHMI Fellow of the Jane Coffin Childs Fund

Department of Molecular and Cellular Biology, Harvard University

Sponsor: Catherine Dulac

Animals rely on instinctive behaviors and homeostatic responses, such as parenting, feeding, mating, and sleeping, to ensure individual and species survival. Maximizing survival requires meeting the most pressing needs at the right time, forcing animals to establish behavioral priorities based on a hierarchy of needs. Neurons controlling many of these behaviors are located within the highly interconnected medial preoptic area of the hypothalamus (MPA), making this structure a likely control hub underlying behavioral hierarchy. However, the neural logic of intra-MPA connectivity and how this directs behavioral priorities across physiological states is unknown.

Using the mouse MPA as a model system, I am studying the cell type-specific structural and functional connectivity underlying key competing behavioral and physiological responses. Further, I am determining how animal states, such as virgin or parent, alter neuron function to induce new behavioral priorities. This work will provide the first depiction of a neural basis of the hierarchy of needs and open new avenues for understanding the neural basis of behavior.

Sarah Innes-Gold Merck and Company Fellow

Department of Chemistry and Chemical Biology, Harvard University

Sponsor: Adam Cohen

Cancer metastasis and immune cell migration require motile movement, meaning the cell membrane must slip relative to the cytoskeleton. Thus, membrane-cytoskeleton attachments in motile cells likely rearrange, allowing tension to propagate across the membrane. In the current literature, there are ~106-fold discrepancies in reported timescales of membrane tension propagation. I hypothesize these discrepancies reflect variability between cell types, arising from differences in membrane microstructure. I specifically hypothesize that in motile cells, transmembrane proteins are arranged to allow membrane flow, enabling rapid tension equilibration, while non-motile cell membranes are structured to impede tension propagation.

I will directly measure tension propagation timescales in motile and non-motile cells and simultaneously characterize the arrangement of cytoskeleton-anchored transmembrane proteins. I will use optical tweezers to stretch membrane tethers, perturbing and measuring tension. I will visualize immobile transmembrane proteins with targeted photochemical labeling and high-resolution fluorescence imaging, revealing how transmembrane protein arrangement regulates membrane fluidity, and how cancer cells might exploit this to metastasize

Aakanksha Jain Jane Coffin Childs Fellow

Department of Neurobiology, Boston Childrens Hosptial

Sponsor: Clifford Woolf

Advancements in vaccine design and immunotherapy have helped us gain insights into how to promote immunity against infections or cancers. However, excessive inflammation associated with immunotherapies, autoimmune diseases, non-healing wounds and even COVID19 is currently at the center of healthcare challenges. Following an inflammatory insult, such as an injury or pathogen invasion, immune cells in the tissues are crucial to resolve inflammation and regain healthy tissue function. Damaging inflammatory signals also activate nearby high threshold sensory neurons– the nociceptors – which are responsible for initiating pain and guarding/withdrawal responses which is believed to prevent further tissue damage. While it is conceivable that nociceptors can cooperate with immune to promote healing, the role of these neurons in shaping the healthy immune landscape of barrier tissues is currently unexplored. In the Woolf lab, I aim to determine the role of nociceptor sensory neurons in restoring the healthy immune profile of barrier tissues following an adverse and painful inflammatory event and develop novel strategies to manipulate neuroimmune interactions using genetic and pharmacological methods. Traditionally, inflammatory conditions are treated with broad immunosuppressants that put the patients at risk for further infections. The ability to fine tune immune function by controlling specific neuronal signals will offer a safer and effective therapeutic strategy for various inflammatory diseases as well as malignancies.

Harris Kaplan HHMI Fellow

Department of Molecular and Cellular Biology, Harvard University

Sponsor: Catherine Dulac

Mammalian social behaviors change dramatically over the lifespan: infants rely on their mothers for food and warmth, adolescents engage each other in social play, and adults mate and parent. This highly conserved social niche trajectory consists of dynamic motivational drives and behavioral repertoires and co-occurs alongside rapid changes in brain organization. However, it remains unclear how developmental changes in behavior result from transformations of the underlying brain circuits.
As a postdoctoral fellow in Catherine Dulac’s lab, I am dissecting these developmental transitions in mammalian brain and behavior. Focusing on the mouse hypothalamus, I am charting the coordinated emergence of transcriptional cell-type identities, spontaneous and stimulus-evoked neuronal activity patterns, and corresponding changes in behavior. Further, I am exploring the robustness and plasticity of these trajectories by manipulating the animal’s sensory and social rearing environment. This work will provide novel insights into the developmental processes that build animal behavior.

Justus Kebschull Jane Coffin Childs

Department of Biology, Stanford University

Sponsor: Liqun Luo

Brain circuits and the neuronal cell types that form them are not static over evolutionary time. Rather, they cause and reflect the changing repertoire of animal behavior. How circuits and cell types change from their ancestral state to support novel behaviors during evolution, therefore, gives us important clues as to their current function. In my project, I will investigate the interaction between the cerebellum and the rest of the brain from this evolutionary angle by studying the progressive expansion and elaboration of the deep cerebellar nuclei, the output pathway of the cerebellum. I will profile transcriptional and projectional cell types of the DCN across species to probe changes in the DCN over deep evolutionary time. I will then integrate this dataset with developmental trajectories of the identified cell types in mouse, to provide mechanistic insight into how brain regions specialize on the level of single cells and circuit wiring to support new functions over the course of evolution.

Anna Kotrys HHMI Fellow

Department of Molecular Biology, Massachusetts General Hospital

Sponsor: Vamsi Mootha

Description: Mitochondria are present in nearly all human cells where they play key roles in energy metabolism, biosynthesis, signaling, and cell death. Mitochondrial homeostasis depends on the proper maintenance and expression of the mitochondrial genome (mtDNA). Germline mtDNA mutations can lead to severe, maternally inherited disorders with limited treatment possibilities. Moreover, somatic mtDNA mutations accumulate in neurodegeneration, cancer and aging. mtDNA is a high copy number genome and a mixture of wild-type and mutant mtDNA molecules can co-exist within one cell resulting in “heteroplasmy”. Heteroplasmy dynamics are governed by a complex mix of random drift and selection, but the underlying molecular mechanisms remain unknown. The aim of my post-doctoral research is to uncover the molecular mechanisms that govern mtDNA heteroplasmy. Mechanistic studies of heteroplasmy dynamics will shed the light on the mitochondrial contribution to human health and disease and possibly inspire novel therapeutic approaches to mtDNA disease.

Andrea Kriz HHMI Fellow of the Jane Coffin Childs Fund

Department of Genetics and Genomics, Boston Children's Hospital

Sponsor: Christopher Walsh

While somatic mutations have been heavily studied in tumors, their prevalence and significance to disease risk in healthy individuals is much less well-understood. The Walsh lab and others revealed that somatic mutation is a widespread phenomenon. Human neurons each contain 100 or more clonal somatic single nucleotide variants (sSNV) at birth, acquired during prenatal development, and gain 15-20 additional sSNVs arising per year. Most somatic variants, including those associated with cancer risk, occur in noncoding regions such as enhancers. Despite being the main source of genetic diversity between cells within an individual, the mechanisms by which noncoding somatic mutations form as well as their functional impact are not well understood. My research will focus on developing new strategies to detect rare noncoding somatic variants as well as dissect their epigenomic impact across different cell types in the human brain. This will help illuminate how much this source of variation contributes to cancer risk and brain disease.​

Benjamin Larson Merck Fellow

Department of Biochemistry and Biophysics, University of California, San Francisco

Sponsor: Marshall Wallace

I am interested in understanding how cells control shape and movement to thrive in different environments. Although often regarded as simple building blocks, single cells frequently execute surprisingly complex, even animal-like behaviors, which are necessary for proper cellular function. In cells, these behaviors emerge from the joint action of myriad molecular components and interactions between the cell and its environment. How this occurs is poorly understood. To better understand and predict cell behavior, I am working to uncover general principles by studying the coordination of walking in a unicellular organism, the ciliate Euplotes.
How can a single cell, lacking a nervous system, coordinate a gait? While unusual in some ways, Euplotes locomotion is amenable to rigorous behavioral analysis, and many underlying cellular processes and molecular components are deeply conserved among eukaryotes. My work combines theory from computer science and non-equilibrium statistical physics with quantitative microscopy experiments to uncover the mechanisms by which Euplotes coordinates its gait and will develop new theoretical and experimental tools for interrogating the control of complex cellular behaviors.

Rosalie Lawrence HHMI Fellow

Department of Biochemistry and Biophysics, University of California, San Francisco

Sponsor: Peter Walter

All cells and organisms mount stress response programs in response to external insults; some recover to baseline after stress, while others suffer from side-effects such as chronically altered proteomes that can reduce cellular and organismal fitness. I study the cellular machinery that executes the Integrative Stress Response (ISR), a highly conserved cellular program that rewires translation in the wake of stresses such as nutrient deprivation, viral infection, or redox imbalance. I seek to understand how the ISR machinery remains flexible enough to both respond to diverse stresses and return to baseline, and how dysregulation of the ISR leads to chronic inflammation and memory disorders in higher organisms. I am particularly excited to leverage recent advances in structural biology to go beyond a static understanding and toward uncovering dynamic conformational transitions in cellular ISR machinery that enable nuanced decision-making. To this end, I use hydrogen deuterium exchange, biochemical and cellular assays, and live imaging to study the key ISR actuator eIF2B both in vitro and in cells.

Chip Le Jane Coffin Childs

Department of Chemistry and Chemical Biology, Harvard University

Sponsor: Emily Balskus

Catechol dehydroxylation is a highly relevant metabolism in the human gut microbiota with a significant impact on human health. A wide range of neurotransmitters, dietary compounds, and drug molecules have been identified as substrates for this uniquely microbial transformation. However, the ability to predict and manipulate such an important process has been hindered by the limited understanding of enzymes that facilitate the transformation. The Balskus group recently identified dopamine dehydroxylase (Dadh) as the enzyme responsible for the conversion of dopamine to m-tyramine in the gut microbiota. Phylogenetic analysis showed that Dadh and its homologs form a unique DMSO-reductase subfamily. These proteins have not been characterized, and the mechanism has not been deciphered. Moreover, a survey of the human gut microbiome revealed a large number of molybdopterin-dependent enzymes with unknown chemical capability. The main focus of my work is to investigate human gut catechol dehydroxylases via a substrate-guided approach. This work will be accomplished by (1) deciphering the structure and mechanism of dopamine dehydroxylase, (2) biochemically characterizing and comparing reactivity of catechol dehydroxylase homologs, and (3) exploring additional molecular scaffolds that could be susceptible to dehydroxylation by unknown molybdopterin dehydroxylases.

Joyce Lee HHMI Fellow of the Jane Coffin Childs Fund

Department of Pediatric Oncology, Dana-Farber Cancer Institute

Sponsor: David Pellman

Common fragile sites (CFSs) are hotspots for genomic rearrangements in cancers, but why and how these rearrangements occur is poorly understood. CFSs are characteristically difficult to replicate and often persist as under-replicated DNA into mitosis. Previous work from my host laboratories showed that stalled DNA replication forks are disassembled upon exposure to the mitotic kinase Cyclin B-CDK1. Unloading of the replisome leads to the formation of DNA breaks at the stalled forks followed by break-end ligation events. Coordinated DNA repair events at converging stalled forks in mitosis could lead to the formation of deletions and sister chromatid exchanges, both of which are signatures of CFS expression.

Recent studies have shown that CIP2A is a mitosis-specific repair factor that localizes to sites of DNA damage and replication stress. My aim is to investigate the role of CIP2A in cellular responses to unreplicated DNA in mitosis and how these processes contribute to genomic instability at CFSs. I will use the Xenopus egg extract system to uncover the biochemical mechanism of CIP2A function and conduct cell-based studies to observe the effects of CIP2A on genome stability.

Jiefu Li HHMI Fellow

Institute for Immunity, Transplantation and Infection, Stanford University

Sponsor: Mark Davis

T lymphocytes are central players of our adaptive immune system for fighting against pathogens as well as aberrant self cells. The recognition, action, and modulation of T cells rely on diverse molecules on their surface, including T cell antigen receptors (TCRs) and numerous signaling modulators, such as CTLA-4 and PD-1. Using systems approaches, I study cell-surface signaling of human T cells, with two focuses: 1) I combine TCR repertoire profiling, computational analysis, and scalable antigen screen to quantify TCR repertoire dynamics in infectious diseases and search for population-shared antigens to inspire vaccine development; 2) I build novel tools for spatiotemporally-resolved quantitative proteomics to determine how the T cell surface proteome evolves under distinct cellular states and look for molecular targets for invigorating or modulating T cell activities.

Gen Li Jane Coffin Childs Fund Fellow

Department of Chemistry, University of California, Berkeley

Sponsor: Christopher Chang

It has long been a focus for the identification of protein drivers and for the development of corresponding cancer therapeutics. Traditionally, drug discovery efforts mainly rely on “druggable” proteins, which possess easily identifiable binding pockets or catalytic active sites. However, over 85% of the proteome is still considered “undruggable”, posing additional challenges for further development. Recently, Activity-Based Protein Profiling, has arisen to spotlight the undruggable proteome via covalent linkage of reactivity-based chemical probes and “ligandable hotspots” in the proteome. In the proposed research, I aim to develop new chemoproteomic platforms based on inexpensive and biocompatible main-group molecules for chemoselective methionine and methionine sulfoxide bioconjugation, to further promote cancer drug discovery. This contribution will be significant because it can develop a ligandability map against undruggable proteome, serve as efficient tools in cancer cell early-stage diagnosis, and further provide a handle to decipher and drug methionine redox regulation in cancer cells, thus yielding novel therapeutics.

Chi-Yun Lin Jane Coffin Childs Fellow

Department of Chemistry, Pennsylvania State University

Sponsor: Joseph Bollinger

Protein-based radicals participate in biological processes and natural product biosynthesis that link to life and death in organisms. One remarkable example is class I ribonucleotide reductases (RNRs), which catalyze DNA synthesis with tyrosyl radical relays. To compete for available resources, particularly in pathogens that live in the context of a host, RNRs have evolved distinct cofactors, assembly strategies, and radical translocation mechanisms. Understanding these distinctions from human counterparts is a key step in developing successful anticancer, antimicrobial, and antiviral drugs that inhibit RNRs. However, tyrosines are abundant and form highly cooperative networks, presenting difficulties in isolating their contribution to vectorial redox. I aim to dissect these tyrosines in the newly discovered class I RNRs to probe the free energy landscape of their one-electron oxidation and determine the active state structures. To further advance the field of redox enzyme design for difficult chemical reactions, I will elucidate the crucial protein environmental factors that modulate productive tyrosyl radical relays and prevent detrimental side reactions.

Zachery Lonergan Jane Coffin Childs

Department of Biology & Biological Engineering, California Institute of Technology

Sponsor: Dianne Newman

Energy conservation is an organizing principle for microbial communities. This conservation becomes challenging for bacterial pathogens that must overcome the host immune response. Nonetheless, bacterial infections are major sources of morbidity and mortality, demonstrating that mechanisms exist for pathogens to persist within hosts. Within the lungs of immunocompromised individuals, immune cells are recruited to eliminate pathogens, but this recruitment is unable to clear the infection. Extreme oxygen gradients exist within the lung environment that require metabolic flexibility for bacterial pathogens to survive. While the unique metabolic sources and requirements for microbes within the lungs is not well-defined, we predict that nitrogen oxides serve an important role in supporting bacterial lung persistence. To test this hypothesis, we are implementing geochemical-based strategies to track bacterial nitrogen oxide metabolism, which will provide new conceptual and technical handles on pathogen activities within the human host.

Chin San (Eric) Loo Merck Fellow of the Jane Coffin Childs Fund

Department of Immunology, Harvard Medical School

Sponsor: Jun R Huh

The human gastrointestinal tract harbors trillions of microbes that have been coevolving with humans for a long time. Growing evidence suggested that the gut microbiota produces a myriad of metabolites, and some of these small molecules possess bioactivity that can shape host development and fitness, such as modulating gut immune cells and promoting brain development. G protein-coupled receptors (GPCRs) represent the largest class of membrane receptors that relay extracellular cues into a cellular response. Many of these GPCRs including orphan GPCRs may evolutionally be designed for communicating with microbes through microbial metabolites. My research seeks to develop a genetic tool and platform that can characterize ligand-activated GPCRs in vivo and uncover GPCRs that sense microbial metabolites. This work potentially sheds light to understand the underlying mechanisms of host-microbiota interaction.

Claire Magnani Merck Fellow of the Jane Coffin Childs Fund

Department of Chemical Biology and Therapeutic Science,The Broad Institute of MIT and Harvard

Sponsor: Stuart Schreiber

There is an urgent need for cancer therapeutics with improved target specificity and novel mechanisms of action. Molecular glues combine the demonstrated capabilities of small molecules as potent drugs with the power of chemically induced proximity. In enabling new protein-protein associations, molecular glues can address many shortcomings of current small molecule cancer therapeutics, which are often limited to protein targets presenting a clear binding pocket. Furthermore, since protein-protein interactions are widely known to facilitate a range of fundamental cellular activities, chemical compounds which intercede on these pathways can provide access to novel mechanisms of action and enhanced target specificity. The wide-ranging therapeutic potential of molecular glue has already been recognized; however, to date the discovery of these compounds has been limited to serendipity or the synthesis of bifunctional molecules. The systematic and generalizable path to molecular glue discovery has not yet been established. In my research, I will leverage the recording and reporting power of DNA encoded libraries to deliver a new path to molecular glue discovery

Venkata S Mandala HHMI Fellow

Laboratory of Molecular Neurobiology and Biophysics, The Rockefeller Universtiy

Sponsor: Roderick MacKinnon

Many transmembrane proteins reside in functionally important clusters on cell membranes. Fluorescence microscopy of membrane proteins in cells has revealed ‘hot spots’ of co-localized proteins such as a_2A-adreneregic G-protein coupled receptors and G proteins participating in signaling complexes. Yet the functional significance of these signaling clusters in cells is not well established. Developing tools to induce controlled clustering of membrane proteins in the lab would thus provide valuable insight into the function of these signaling complexes in cells.

My project proposes three complementary strategies to induce controlled protein clustering in lipid bilayers. The approaches span raft-forming lipid mixtures, tetraspanin and MARVEL domain 4-TM proteins, and membrane-anchored scaffolding proteins with multiple PDZ domains. These tools will be applied to a signaling pathway comprised of G protein-gated K⁺ channels (GIRK) and their activator, the βγ complex of G proteins (Gβγ). The extent of protein clustering and the subsequent effect on activity will be assessed using fluorescence microscopy and electrophysiology

Elija Mena HHMI Fellow

Department of Genetics, Brigham & Women's Hospital

Sponsor: Stephen J Elledge

Ubiquitylation is a post-translational modification that regulates the stability of thousands of proteins in our cells. The specificity for ubiquitylation is typically conferred by E3 ubiquitin ligases that attach ubiquitin onto substrate proteins. Despite the critical role that ubiquitylation plays in regulating the abundance and activity of many proteins, most ubiquitylation pathways are still poorly understood and many of the estimated ~600 E3 ubiquitin ligases have no known protein substrates.

Our lab has developed the Global Protein Stability (GPS) assay, which is a way to rapidly monitor protein stability using fluorescent proteins. We have recently been adapting this approach for library-on-library genetic screens in order to map, in parallel, dozens of ubiquitylation substrates to their cognate E3 ubiquitin ligases. We have also been using GPS screens to find degradation pathways specific to particular intracellular compartments. Together, these approaches will shed light on ubiquitylation pathways that are important for human health.

Carlos Mendez-Dorantez Jane Coffin Childs Fellow

Department of Oncologic Pathology, Dana Farber Cancer Institute

Sponsor: Kathleen Burns

Long interspersed element-1 (LINE-1) is the only active, protein-coding transposon in humans. LINE-1 overexpression and LINE-1 retrotransposition are hallmarks of human cancers, although the impact of LINE-1 activity on cancer genomes and cancer cell growth remains poorly understood. My research focuses on addressing the hypothesis that LINE-1 retrotransposition causes substantial gross genome instability in cancers. Supporting this hypothesis, a recent pan-cancer analysis demonstrated associations between somatically-acquired LINE-1 insertions and segmental copy-number changes. Moreover, our lab recently identified that the Fanconi anemia/ BRCA pathway is required for growth of LINE-1(+) cells, suggesting that this DNA repair pathway might limit genotoxic effects of LINE-1. I am developing several approaches to assess the impact of LINE-1 on genome integrity, and I am evaluating the contribution of the FA/ BRCA pathway to LINE-1-associated DNA damage. These studies will be the first to evaluate the scope of LINE-1-mediated genome instability and should inform efforts to exploit LINE-1 genotoxicity as a cancer therapeutic strategy.

Phi Nguyen Jane Coffin Childs Fund Fellow

Department of Psychiatry, Columbia University/New York State Psychiatric

Sponsor: Rene Hen

Major depressive disorder (MDD) is a psychiatric disorder with a lifetime prevalence of ~15% and is the leading cause of disability worldwide¹. The societal burden of MDD is immense, causing profound personal suffering and economic loss, which has recently been intensified by the Covid-19 pandemic². The most effective treatments for MDD, a class of antidepressants called the selective serotonin reuptake inhibitors (SSRIs), are successful in achieving remission, but only in ~40% of patients³. Despite being in use for over 50 years, it remains unknown how SSRIs modulate neural circuit function in patients that achieve remission and where these mechanisms are disrupted in those that do not. Thus, a fundamental question remains: What are cellular and molecular mechanisms that mediate antidepressant response and resistance? Defining the answers to this question could provide fundamental insights into the pathophysiology of MDD and uncover novel substrates for future precision medicine approaches.

Yitzhak Norman Jane Coffin Childs Fund Fellow

Department of Neurological Surgery, University of California, San Francisco

Sponsor: Edward Chang

Speech is a defining characteristic of human cognition. It provides humans with the flexibility to convey an unlimited range of thoughts and feelings using a limited number of basic elements. Over the past decade, intracranial electrocorticography (ECOG) recordings in patients have provided invaluable insights into the neural mechanisms underlying speech perception and production. While significant progress has been made, basic questions still remain regarding the functional architecture of the neuronal circuits involved. Particularly, we do not know how the brain assembles phonemes into words, and words into meaningful goal-directed utterances. Such phonemic-to-semantic transformation relies on real-time interactions between the speech cortex and distributed memory networks that encode, store, and retrieve our lexical and semantic knowledge quickly and efficiently. The hippocampus, as a critical node in this declarative memory system, is believed to play a key role in coordinating such processes in real time.

My research seeks to elucidate the cortical-hippocampal interaction during speech perception and production, and more broadly, to unravel the interface between speech representations and long-term memory. To accomplish this, I combine ECOG recordings with 7T fMRI to measure neuronal activity simultaneously from the hippocampus and speech cortex during perception and production of speech.

Xingjie Pan HHMI Fellow of the Jane Coffin Childs Fund

Department of Chemistry and Chemical Biology, Harvard University

Sponsor: iowei Zhuang

During mammalian development, coordinated cell differentiation and migration convert a simple neural tube into a brain with more than a hundred anatomical regions and probably more than a thousand cell types. How do these cell types emerge? How do cells migrate to their destined locations? How do cells communicate with each other? These are some fundamental problems in brain development.

As a postdoctoral fellow in Xiaowei Zhuang’s lab at Harvard, I develop new methods to systematically study these problems in mouse brain development. I develop new computational methods to connect cells from MERFISH spatial transcriptomics measurements into trajectories and determine cell-cell communication pathways activated in each cell. The reconstructed trajectories will allow me to comprehensively map the differentiation, maturation, and migration of individual cells. I will identify which cell-cell communication pathways are functionally crucial for generating each cell type. Then I will develop high throughput imaging-based screen methods to validate the discoveries.

Julia Rogers Jane Coffin Childs Fund Fellow

Department of Systems Biology, Columbia University

Sponsor: Mohammad Al Quraishi

Cells efficiently convert environmental information into specific functional responses through cascades of biochemical reactions and biomolecular interactions. High fidelity signal transduction requires spatiotemporal regulation of these molecular events. This can be accomplished through phase separation. Many signaling condensates dynamically assemble through multivalent protein–protein interactions mediated by modular interaction domains. How the molecular factors that drive phase separation enable coordinated and precise flow of information among myriad signaling pathways remains a mystery. To answer such questions that encompass molecular- and systems-level phenomena, my research focuses on developing integrative data- and physics-based modeling frameworks using the tools of machine learning and statistical mechanics. Using these approaches, I aim to decipher the modular grammar of signaling proteins that governs phase separation and, more broadly, the biophysical principles that underlie cell homeostasis.

Taehyun Ryu Jane Coffin Childs

Department of Genetics, Harvard Medical School

Sponsor: Chao-ting Wu

Ultraconserved elements (UCEs) are a set of DNA sequences that exhibit perfect conservation across the genomes. I learned of UCEs and their putative role in maintaining genome integrity at a seminar by Dr. Chao-ting Wu. Scattered across genomes, unique, and 200bps or greater in length, UCEs have remained unchanged for over 300 million years. Yet, their extreme sequence conservation is still a mystery. Although my Ph.D. training is in the DNA repair field, I decided to join Dr. Chao-ting’s lab as a postdoctoral researcher and explore the biology of UCEs. Previous studies have demonstrated that UCEs can contain transcription factor binding motifs an function as enhancers to regulate tissue-specific transcription. However, no regulatory or proteincoding functions can explain such extreme sequence conservation. My research will focus on testing a model that can explicitly address such an explanation. I hypothesize that homologous UCEs compare their sequences via pairing and any detected discrepancies in sequence or copy number will lead to cell death and/or disease onset. As a result, genome integrity would be maintained by culling out cells carrying deleterious rearrangements. I will assay this model with different approaches – a) computational analyses, b) CRISPR-based genome editing, and c) imaging techniques. Ultimately, the potential of UCEs to sense and cull deleterious rearrangements genome-wide offers a unique yet intriguing and still largely unexplored potential general strategy for treating diseases derived from rearrangements, regardless of the etiology of diseases.

Gavin Schlissel Jane Coffin Childs

Department of Biology, Whitehead Institute for Biomedical Research

Sponsor: Pulin Li

Animals rely on effective coordination of cell behavior in all phases of their development and lifespan. Cells communicate to coordinate their activity using several physical or chemical communication strategies, which are often interdependent. One communication strategy central in development and in adult animals relies on secreted signaling proteins that bind membrane-tethered receptors in diverse target tissues to affect cell identity or behavior. Whereas we understand in great detail how signals are synthesized, secreted, received and processed, we understand comparatively very little about how signals travel from their origin to their destination. I use molecular genetic and synthetic biology tools in cultured mammalian cells to reconstitute cell signaling events, and I use these reconstituted signaling pathways to understand how secreted protein signals navigate the extracellular environment in developing or adult tissues.

Jiarui Song HHMI Fellow of the Jane Coffn Childs Fund

Department of Biochemistry, University of Colorado, Boulder

Sponsor: Thomas Cech

Histone methyltransferase PRC2 (Polycomb Repressive Complex 2) silences genes via successively attaching three methyl groups to lysine 27 of histone H3 (H3K27me3). Several research groups including ours demonstrated that PRC2 associates with numerous pre-mRNA and lncRNA transcripts with a strong binding preference for G-quadruplex forming RNA. However, the structural details of their interactions have so far been unclear. My research provides a 3.3Å-resolution cryo-EM structure of a PRC2-RNA ribonucleoprotein complex. Notably, G-quadruplex RNA bridges the dimerization of PRC2 with a symmetric interface comprised of two copies of the PRC2 catalytic subunit EZH2. Especially, EZH2 SET domain is indicated to directly facilitate the RNA-mediated dimerization of PRC2. Interestingly, those residues were previously characterized in the PRC2-nucleosome cryo-EM structure to physically interact with the histone H3 tail and nucleosome DNA. Therefore, I hypothesize that in the dimerized PRC2-RNA complex, RNA inhibits PRC2 activity by limiting H3 tail accessibility to the active site. Overall, my study provides a new perspective of RNA regulation of chromatin modifiers.

Elizabeth Stone Jane Coffin Childs Fund Fellow

Department of Chemistry, University of California, Berkeley

Sponsor: Michelle Chang

The discovery of novel antitumor drugs requires the development of new methods to synthesize molecules of increasing diversity and complexity to meet the challenges of drug efficacy and safety. Biocatalysis provides an attractive strategy to perform chemical reactions under mild and sustainable conditions. The Chang Lab has recently discovered a family of radical halogenases that perform the regio- and stereoselective chlorination of unactivated, aliphatic C–H bonds within several amino acid substrates. Despite the synthetic utility of organohalides, there are limited biosynthetic and chemical methods for the selective chlorination of unfunctionalized alkanes beyond this example.

Using mechanistically-guided protein engineering, my research aims to expand the substrate and reaction scope of these enzymes to produce noncanonical amino acids bearing versatile functional group handles, including halogens or azide. These synthetic residues will then be incorporate into biological molecules of interest, such as known anticancer peptides, and can be further functionalized to access diverse, cyclic structures. Overall, this strategy provides a fully biosynthetic method for producing novel analogs of anticancer peptides with the goal of discovering improved drugs.

Xiaochen Sun Jane Coffin Childs Fund Fellow

Department of Biology and Applied Physics, Stanford University

Sponsor: March Schnitzer

The ability to learn and memorize is essential for all living organisms to adapt to the ever-changing environment, and it serves as the foundation for higher-order cognitive processes such as reasoning, planning, and decision-making. However, the neuronal basis of memory remains unclear — it is largely unknown how memory-related information is represented by populations of neurons in the brain, and how that representation is formed as a result of learning-induced plasticity. By studying the activity of neurons underlying long-term memory using in vivo imaging and opto/chemogenetics, I hope to understand the neural mechanisms by which new information is learned and processed in neuronal populations. This work will provide insights into the computational principles that govern learning in biological and artificial neural networks.

Marla Tharp HHMI Fellow

Whitehead Institute of Biomedical Research

Sponsor: David Page - Co-Sponsor Rudolph Jaenisch

Cancer affects men and women differently. For example, glioblastoma, the most aggressive form of brain cancer, has a male-biased incidence rate and poorer response to standard treatments in men versus women. My research investigates the genetic and molecular basis of sex differences in glioblastoma from the perspective of microglia, the resident immune cells of the brain. Microglia are a major player in the brain tumor microenvironment and promote tumor growth and metastasis. Using XX and XY human microglia isolated from healthy brain regions and brain tumors, I am identifying sex-biased genes and biological pathways that are responsible for establishing sexually dimorphic brain tumor microenvironments. Further, I am testing how possessing an XX or XY sex chromosome complement drives the observed genome-wide sex-biased gene expression patterns in microglia, in particular, through X-linked genes that aberrantly escape X chromosome inactivation or homologous X-Y gene pairs with imbalanced expression or function. I anticipate that my research will lay the groundwork for more effective and sex-specific treatments for glioblastoma.

Jenelle Wallace Jane Coffin Childs Fund Fellows

Department of Neurobiology, University of California, San Francisco

Sponsor: Alexander Pollen

Over the last six million years, the human brain has evolved in remarkable ways. The story of human brain evolution is written in our genome, but determining which genetic changes are responsible for cognitive differences between humans and other primates is challenging. Differences in gene expression are likely controlled by regions of the genome that were once considered “junk DNA” but are in fact crucial regulators. Utilizing stem cell-derived neurons from humans and other primates, I will investigate how alterations in the genetic sequences of these regulatory elements have driven human brain evolution.

Hannah Wayment-Steele Jane Coffin Childs Fellow

Department of Biological Chemistry and Molecular Pharmacology, Brandeis University

Sponsor: William Shih and Dorothee Kern

Deep learning methods have revolutionized structural biology by accurately predicting single structures of proteins and protein-protein complexes. However, biological function is rooted in a protein’s ability to sample different conformational substates, and disease-causing point mutations are often due to population changes of these substates. This has sparked immense interest in expanding the capability of algorithms such as AlphaFold2 (AF2) to predict conformational substates. We demonstrate that clustering an input multiple sequence alignment (MSA) by sequence similarity enables AF2 to sample alternate states of known metamorphic proteins, including the circadian rhythm protein KaiB, the transcription factor RfaH, and the spindle checkpoint protein Mad2, and score these states with high confidence. Moreover, we use AF2 to identify a minimal set of two point mutations predicted to switch KaiB between its two states. Finally, we used our clustering method, AF-cluster, to screen for alternate states in protein families without known fold-switching, and identified a putative alternate state for the oxidoreductase DsbE. Similarly to KaiB, DsbE is predicted to switch between a thioredoxin-like fold and a novel fold. This prediction is the subject of ongoing experimental testing. Further development of such bioinformatic methods in tandem with experiments will likely have profound impact on predicting protein energy landscapes, essential for shedding light into biological function.

Caleb Weinreb Jane Coffin Childs

Department of Neurobiology, Harvard Medical School

Sponsor: Sandeep Datta

I study the deep statistical structure of behavior to learn how it is shaped by ongoing brain activity. The purpose of the central nervous system is to coordinate an animal’s actions in space and time. The power of mammalian brains is evident in the variety and expressiveness of their behavior, yet it is precisely these qualities that make the behavior difficult to annotate and record – steps that are prerequisite for modern data analysis. As a consequence, neuroscience has mostly been limited to a narrow set of behaviors and well-defined tasks. This limitation is especially severe for the study of social behavior, in which the spontaneous actions and reactions of two interacting animals created an added level of complexity.

Recently, the advent of new tools in machine learning have made it possible to quantify behavior with much greater precision and richness. My research focuses on creating new tools for behavior measurement and applying them to rodent social behavior, with the specific goal of understanding how social interaction is shaped by the prefrontal cortex.

Katherine Xue Jane Coffin Childs

Department of Biology and Medicine, Stanford University

Sponsor: Dimitri Petrov and David Relman

The trillions of microbes that live in and on the human body play key roles in health and disease. However, little is known about how microbes evolve in complex communities, even though this evolution can have important consequences for human health. I will study how adaptation and dispersal drive the evolution of antibiotic resistance in microbial communities, both in the human gut microbiome (in vivo) and in experimental, gut-derived microbial communities (ex vivo). First, I will track evolution in the human gut microbiome in a cohort of healthy individuals treated with ciprofloxacin. Using strain-resolved metagenomic sequencing, I will identify selective sweeps and strain replacements to determine how natural microbial communities evolve in response to a disturbance. Next, I will examine how adaptation and dispersal shape the evolution of gut-derived microbial metacommunities. These experimental metacommunities allow me to test how dispersal shapes the rates and mechanisms of adaptation in more controlled, laboratory contexts. Finally, I will study adaptation and transmission in the human gut microbiome by tracking strain transmission in cohabiting individuals before and after antibiotic treatment. This work will combine new computational and experimental approaches to shed light on how microbial communities evolve in the context of human health.

Liewei Yan Jane Coffin Childs Fund Fellow

Department of Biology, Carnegie Institution of Washington

Sponsor: Kamena Kostova

Ribosomes are complex molecular machines that translate mRNAs into proteins and are essential for sustaining life. While the ribosome functions in cellular environments that are markedly diverse, its composition has traditionally been seen as static after assembly. Exciting new studies challenge this concept and provide evidence that organisms assemble different types of ribosomes during development, stress response, or disease. For example, during embryogenesis, zebrafish assemble two types of ribosomes with distinct structures: maternal and somatic. Although this ribosome heterogeneity is predicted to alter protein synthesis, no experimental evidence yet exists to demonstrate this. I will use a multidisciplinary approach to test how changes in ribosome composition affect translation during zebrafish development.