QuantBio/NSF Simons Poster Session

Date:

Thursday, May 23, 2019, 4:00pm to 5:30pm

Location:

Maxwell Dworkin Lobby

Poster Summary

#	Author	Institution	Poster Title
1	Jamilla Akhund-Zade	Harvard University	Characterizing evolutionary strategies in wild Drosophila thermal preference via high resolution temporal sampling and broad geographic collections
2	Ashish B. George	Boston University	Survival of the chiral: Chirality provides a direct fitness advantage and facilitates intermixing in cellular aggregates
3	Felix Barber	Harvard GSAS	How do they know? Probing cell size regulation in budding yeast
4	Katharine Best	Faculty of Arts and Sciences	Modelling dose dependent immune responses in acute Zika infection
5	Silvia Canas Duarte	Harvard University, GSAS	Understanding rare events in bacteria: Revisiting R1 Plasmids losses
6	Shivani Chhabra	University of Massachusetts Medical School	Quantifying the regulatory role of individual Transcription Factors in Escherichia coli
7	Tim Chiang	Harvard SEAS	RNA Selectivity in Viral Assembly
8	Zachary Chiang	Harvard University	Genome-wide in situ sequencing
9	Yongmin Cho	Harvard Medical School	Quantitative characterization of animal aging: A comprehensive picture from phenotype to genotype
10	Wenping Cui	Boston University	Diverse communities behave like typical random ecosystems
11	Ivana Cvijovic	Harvard University	High-resolution lineage tracking reveals traveling wave of adaptation in laboratory yeast
12	Margherita De Marzio	Harvard Medical School	Effect of mechanical compression on the gene expression of Human Bronchial Epithelial Cells
13	Rui Fang	Harvard GSAS	Understanding the Functional Dynamics of the 26S Proteasome by Nucleotides-proteasome Interaction Study
14	Travis Gibson	Harvard Medical School	Robust and Scalable Models of Microbiome Dynamics
15	Alexander Golden	Boston University	Tracing Lineages Through Spatial Transitions
16	Anupam Gupta	School of Engineering and Applied Sciences	Motility and adhesion gradient induced vertebrate body axis elongation and somite formation
17	Sabina Haque	Harvard University	Stochasticity and magnetoreception in models of magneto-aerotaxis
18	Matthew Holmes	Harvard University	Single-molecule imaging reveals distinct subcomplexes of the Bacillus subtilis division machinery
19	Tanush Jagdish	Harvard University, Harvard Medical School	Exploring Major Evolutionary Transitions Using Experimental Evolution
20	Ofer Kimchi	GSAS	Developing a synthetic post-translational protein oscillator
21	Julian Kimura	Harvard University	Development of Hofstenia miamia and the Embryonic Origin of Neoblasts
22	Thomas LaBar	Harvard Faculty of Arts and Sciences	Mutation supply and the genetic mechanisms of cellular evolution
23	Anastasia Leshchyk	Worcester Polytechnic Institute	Revealing Tumor Heterogeneity in Triple-negative Breast Cancer at Single-cell Resolution through a De Novo Deep Clustering Approach
24	Ethan Levien	Harvard SEAS	Quantifying phenotypic variability in finite populations
25	Jie Lin	SEAS	Coevolution of multiple growth traits in microbial populations under serial dilution
26	Nathan Lord	Harvard	Co-receptor feedback shapes the Nodal signaling gradient
27	Victor Luria	HMS	Variation and novelty in evolution: novel genes continuously arise, enable protein structural innovation and function in the brain
28	Anastasia Lyulina	Harvard Medical School	Population genomics of the critically endangered spoon-billed sandpiper
29	Elizabeth May	GSAS	Molecular and Cellular Mechanisms of Neuronal Connectivity
30	Harry McNamara	GSAS	Electrical Reaction-Diffusion Pattern Formation in Synthetic Biological Tissues
31	Matthew Melissa	GSAS	Traveling-Wave Models of Evolution: The Generalized Infinitesimal Limit
32	Yuchuan Miao	Harvard Medical School	Feedback control of the human segmentation clock
33	Danai Montalvan	SEAS	A synthetic nucleus for synthetic cells
34	Emily Moore	Harvard School of Dental Medicine	The role of Galphas signaling in postnatal tooth development
35	Maria Mukhina	Harvard	Towards mechanoluminescent force nanosensor for biological systems
36	Anjalika Nande	Harvard University	The Role of Drug Kinetics on the Evolution of Resistance.
37	Maximilian Nguyen	Harvard Medical School	Reciprocity: A Functional Measure of Coupled Transcription Factor Binding
38	Alan Pacheco	Boston University	Generation of microbial community phenotypes through reinforcement learning
39	Vinuselvi Parisutham	UMASS Medical School Worcester	Influence of Chromosomal Position in Controlling Prokaryotic Gene Expression
40	Philip Pearce	Massachusetts Institute of Technology	Flow-induced symmetry breaking in growing bacterial biofilms
41	Rostam Razban	Harvard University	Protein melting temperature and protein folding free energy are not interchangeable in light of protein evolution
42	Annika Rohl	Harvard Medical School	PageRank on Multilayer Networks
43	John Russell	SEAS	Phenotypic Tracking of Microbes with Raman Spectroscopy
44	Carlos Sanchez	Harvard Medical School	Ultraprecise Measurement of Fitness in Bacteria
45	Georgia Squyres	Harvard University	Z ring assembly is regulated by FtsZ filament binding proteins
46	Elad Stolovicki	SEAS	Drop chemostats on a chip
47	Charlotte Strandkvist	Harvard Medical School	Fluctuation-response relations in biological systems
48	Lei Sun	Harvard Medical School	Stochastic gene expression influences the selection of antibiotic resistance mutations
49	Tony Tsai	Harvard Medical School	An adhesion code enables robust pattern formation in the zebrafish spinal cord
50	James Valcourt	GSAS	Understanding lineage commitment and context-dependent response to TGF-B superfamily signals during early mammalian development
51	Debra Van Egeren	Harvard University	Recording transcriptional and cell division histories in individual hematopoietic progenitor cells
52	Suyang Wan	HAVARD	Genome-scale Time-course Genotype-Phenotype Mapping with Promoter-based Barcoding System
53	Shou-Wen Wang	Harvard Medical School	Emergence of collective oscillations in adaptive cells
54	Xin Wang	Harvard Medical School	Overcome Competitive Exclusion in Ecosystems
55	Sean Wilson	Harvard University	The Role and Regulation of Cell Wall Hydrolases in Bacillus subtilis
56	Hao Wu	GSAS	Olfactory Evidence Accumulation in Mice
57	Min Wu	University Of Basel	The genetic mechanism of selfishness and altruism in parent-offspring coadaptation
58	Xingbo Yang	FAS	Biophysical studies of metabolic control in mouse oocytes and embryos
59	Liyuan Zhang	SEAS	Cell in Gel
60	Pengjuan Zu	MIT	Chemical communication explains plant-herbivore interaction networks: an information theory approach

Poster Abstracts

#1
Jamilla Akhund-Zade
Harvard University
Molecules, Cells, and Organisms

Characterizing evolutionary strategies in wild Drosophila thermal preference via high resolution temporal sampling and broad geographic collections

Bet-hedging is an evolutionary strategy in which a single genotype encodes a distribution of phenotypes as a means of ensuring that, even when the environment fluctuates unpredictably, some individuals will be fit. This is in contrast with adaptive tracking, or adaptation via natural selection, where phenotypes are genetically determined and the mean phenotype tracks with the environmentally determined optimum. While there is theoretical evidence that behavioral variability is a result of bet-hedging, there are few studies that test this idea empirically. We previously developed a model in which individual Drosophila thermal preferences affect life history, and in turn lineage evolution (Kain et al., 2015). This model predicts that bet-hedging is advantageous in regions of high seasonality and short breeding seasons, such as Boston, MA. An adaptive-tracking strategy is favored in locations with long and mild breeding seasons, such as Miami, FL. To test the modeling predictions empirically, we collected Drosophila across the U.S. as well as throughout the breeding season. Our hypotheses, deriving from the model, are that flies from northern latitudes will show roughly constant mean preference over the season and low heritability for individual thermal preferences. From our weekly seasonal collections of D. melanogaster in Boston, MA and Charlottesville, VA, we failed to see the fluctuations in mean preference that are predicted by adaptive tracking. A heritability analysis of D. melanogaster isofemale lines from FL, VA, MA, TX, PA, and Southern CA shows that heritability is higher in areas predicted to be adaptive-tracking advantageous and lowest in areas with high bet-hedging advantage, consistent with our hypothesis. A parallel investigation into the molecular mechanisms of thermal preference individuality found that expression differences in trpA1 among isogenic animals explain roughly 20% of the observed thermal preference differences among individuals.

#2
Ashish B. George
Boston University
Physics

Survival of the chiral: Chirality provides a direct fitness advantage and facilitates intermixing in cellular aggregates

Chirality in shape and motility can evolve rapidly in microbes and cancer cells. To determine how chirality affects cell fitness, we developed a model of chiral growth in compact aggregates such as microbial colonies and solid tumors. Our model recapitulates previous experimental findings and shows that mutant cells can invade by increasing their chirality or switching their handedness.
The invasion results either in a takeover or stable coexistence between the mutant and the ancestor depending on their relative chirality. For large chiralities, the coexistence is accompanied by strong intermixing between the cells, while spatial segregation occurs otherwise. We show that the competition within the aggregate is mediated by bulges in regions where the cells with different chiralities meet. The two-way coupling between aggregate shape and natural selection is described by the chiral Kardar-Parisi-Zhang equation coupled to the Burgers' equation with multiplicative noise. We solve for the key features of this theory to explain the origin of selection on chirality. Overall, our work suggests that chirality could be an important ecological trait that mediates competition, invasion, and spatial structure in cellular populations.

#3
Felix Barber
Harvard GSAS
Molecules, Cells and Organisms

How do they know? Probing cell size regulation in budding yeast

Cells across all domains of life display size regulation, coupling their growth and division to constrain the distribution of sizes observed within a cell population. We present preliminary experiments probing the mechanism of size regulation in budding yeast. Our experiments demonstrate that, contrary to widespread belief, cell size regulation in this organism appears to be independent of the dilution of the transcriptional inhibitor Whi5 during the G1 phase. Additionally, we will present recent work determining the influence of different genes involved in the Start transition on size control in budding yeast.

#4
Katharine Best
Faculty of Arts and Sciences
Organismic and Evolutionary Biology

Modelling dose dependent immune responses in acute Zika infection

The key mechanisms of immune control of acute Zika virus infection are not fully understood, and furthering our knowledge of the within host dynamics of Zika virus will be important to the development of effective antiviral strategies. The majority of within host dynamics research has been performed in non-human primate (NHP) models of Zika infection, and understanding the role of inoculum dose is an important component in being able to translate results from a controlled experimental infection to a natural infection. Here we use mathematical modelling to analyze the within host dynamics of Zika virus in NHPs after infection at different inoculum doses. We find strong evidence for innate immune control of plasma viral load and dose dependence in the timing or strength of this immune response.

#5
Silvia Canas Duarte
Harvard University, GSAS
Systems Biology

Understanding rare events in bacteria: Revisiting R1 Plasmids losses

Bacterial plasmids are of great importance for the exchange of genetic material and are responsible for most antibiotic resistance in hospitals. Several low-copy plasmids, such as R1, have been extensively characterized and use several mechanisms to control their copy numbers, segregate copies to both daughters, and encode post-segregational killing (PSK) systems to punish plasmid-free cells. However, despite the in-depth knowledge of the molecules involved in such plasmid stability systems, little is known about their dynamic behavior.
Plasmid loss rates, i.e., the probability that a plasmid-containing cell produces a plasmid-free cell at cell division, have traditionally been indirectly inferred from bulk assays, by monitoring the fraction of plasmid-free cells over time. However, even when such experiments are done carefully, it is virtually impossible to disentangle real losses from the effects of competition between plasmid-containing and plasmid-free cells.
Using an ultra-high throughput microfluidic imaging platform, we can quantify miniR1 plasmid losses in single cells while completely removing the confounding effects from growth competition. We found that miniR1 plasmid is much more stable than previously reported, with intrinsic loss rates of 10-6 or less. By tracking losses along the growth curve from exponential to stationary phase, we further found loss rates go up by several orders of magnitude in the divisions as cells enter stationary phase.
Our ability to directly visualize billions of cell divisions further allow us to monitor the entire history of cells leading up to a loss event, which is particularly important since losses are so rare, and thus could be driven by other rare events, such as highly asymmetric divisions, rare chromosome dynamics, or mutations. We have found evidence that such mutations could be an important contributor to plasmid losses.

#6
Shivani Chhabra
University of Massachusetts Medical School
Program in Systems Biology

Quantifying the regulatory role of individual Transcription Factors in Escherichia coli

Transcription Factors (TF) are typically categorized as “Repressors” or “Activators” depending upon their qualitative impact on gene expression levels. However, these labels can be misleading as they pertain to the net role of a TF on a particular promoter which is often convolved with higher order effects such as TF-TF interactions, feedback or other downstream effects rather than an intrinsic property of the TF itself. In fact, it is not uncommon to find a TF annotated with similar relative binding locations on two different genes but with opposite regulatory effects. Our goal is to characterize the TFs of E. coli individually, on a controlled target gene that is regulated only by the TF in question in order to characterize their fundamental regulatory role, stripped of context. Here, we use a synthetic biology approach to characterize a set of TFs from E. coli and develop an experimental technique to measure the quantitative regulatory role (which we call “regulatory potential”) of a TF when acting alone on a promoter. We characterize the regulation of this set of TFs at two of the most common TF binding locations: directly downstream from the transcription start site and centered at 61 base pairs upstream of the promoter. We demonstrate that the regulatory potential of a TF can be quantitatively measured through a thermodynamic model of gene expression to infer the fundamental regulatory role of each TF as a function of binding location and TF concentration. We find that TFs can show a diverse range of regulatory potentials beyond the qualitative descriptions of activation or repression.

#7
Tim Chiang
Harvard SEAS
Applied Physics

RNA Selectivity in Viral Assembly

The simplest viruses consist of only a single molecule of RNA and multiple copies of a single protein. During an infection, viral RNA is encapsulated by its protein shell with extreme selectivity, the mechanism for which is largely unknown. A proposed mechanism for this selectivity is the specificity of the interactions between the capsid proteins and the RNA, facilitated by the presence of “packaging signals” (PS). Recently, experiments on bacteriophage MS2 have indicated that capsids assemble through a nucleated pathway, which the specific interactions between coat protein and packaging signals could help to facilitate. The faster nucleation kinetics may underlie the mechanism for genome selectivity found in natural viruses. Our experiment utilizes interferometric scattering microscopy to quantify the assembly kinetics of single virus particles, as well as a technique based on DNA oligo binding to knock out sequence effects in wild-type RNA. We have observed a decreased nucleation rate for knock-out and non-cognate RNAs. This study represents a significant step towards understanding the mechanism of RNA selectivity in viral assembly.

#8
Zachary Chiang
Harvard University
Bioinformatics and Integrative Genomics

Genome-wide in situ sequencing

The spatial organization of the genome plays an important role in cell fate and function through control of nuclear processes such as gene regulation. In principle, the nucleus is amenable to study in its native context via high resolution imaging, which can capture many spatial and structural features simultaneously. However, in practice, no tools exist for genome-wide imaging.

We have developed an in situ sequencing method to directly resolve the 3D structure of the genome in its native context within single cells. Here we describe the methodological developments underpinning this new imaging/sequencing approach. In brief, we construct a whole-genome sequencing library in situ in fixed cells via enzymatic fragmentation, adaptor ligation, and amplification of genomic DNA. We then interrogate the amplicons via sequencing by ligation, wherein rounds of sequencing are read out using fluorescence microscopy. This sequencing process is automated using a fluorescence microscope with integrated fluidics controlled by MATLAB.

As a proof of principle, we use this approach to spatially resolve hundreds of whole-genome sequencing reads per cell from hundreds of individual cells. We expect this platform technology will expand the scope of possible measurements of genome organization, including high-throughput genome-wide architectural mapping of higher order chromatin folding and chromosome domains. We anticipate these novel imaging-based genomic measurements will yield new insights about the epigenetic processes that underlie genome structure and regulation.

#9
Yongmin Cho
Harvard Medical School
Systems Biology

Quantitative characterization of animal aging: A comprehensive picture from phenotype to genotype

Aging is the largest risk factor for most chronic human diseases. Understanding aging could enable us should be able to enhance the quality of life. Aging studies in model organisms have demonstrated that aging rates are directly and indirectly influenced by signaling networks controlling fundamental physiological processes such as metabolism, stress, and immune response. In all of these, we can expect a high level of complexity. Therefore, a systems-level understanding should augment the particular efforts on individual mechanisms. There are several model systems that have contributed to our understanding but each has limitations for using systems approaches. To remedy this, the Kirschner lab has begun studies of the small crustacean Daphnia magna, as a model organism for approaches to the aging process. Daphnia promises to be a powerful model organism for several reasons: Its short lifespan (50 days). Its complex behaviors relevant to human healthspan (including easy measurements of movement, neurological response, fecundity and heart rate including its human cardiotoxic drugs). Its main advantage may be the ease of perturbation by small molecules. Daphnia is able to live in small volumes and is permeable to small drugs by diffusion. It is used internationally to test for toxins in the water supply. The key element in any systems-level approach is to define a set of relevant response measures and a set of environmental variables or perturbants and try to infer causation using potential systems elements and then couple them to well-established phenotypic outcomes using modern statistical (machine learning) methods. A related approach has been used in our lab using a set of kinase inhibitors to deconvolve a new noncanonical Wnt signaling pathway (PMID 24707051). I will first establish the physiological metrics of healthspan by monitoring longitudinal physiological changes such as behavior, heart rate and whole-animal level oxygen consumption as well as by probing protein and lipid biomass changes at the tissue level via normalized Raman imaging technique, developed in the Kirschner lab. I will make use of a parallel comprehensive longitudinal study of Daphnia proteomics, phosphoproteiomics, transcriptomics and lipidomics underway. I will also participate in the application of single cell transcriptomics. It will be my goal to combine the age-associated proteomic data with phenotypic data to correlate aging with molecular and behavioral changes. What is missing from this picture is perturbation. It is here where Daphnia can make its most unique contribution by allowing for us to screen drugs (in small volumes using diffusional uptake), against poroteomic/transcriptomic changes, lifespan and healthspan (behavioral phenotypes). Though this is a longer term goal, steps toward realizing it can be taken in this fellowship period. Deconvoluting these disparate functionalities is computationally a quintessential systems approach. One exciting additional outcome could be the identification of new small molecules that delay the loss of phenotypic behavioral and molecular characteristics as the animals age.

#10
Wenping Cui
Boston University
Physics

Diverse communities behave like typical random ecosystems

With a brief letter to Nature in 1972, Robert May triggered a worldwide research program in theoretical ecology and complex systems that continues to this day[1]. Building on powerful mathematical results about large random matrices, he argued that systems with sufficiently large numbers of interacting components are generically unstable. In the ecological context, May’s thesis directly contradicted the longstanding ecological intuition that diversity promotes stability[2–4]. In economics and finance, May’s work helped to consolidate growing concerns about the fragility of an increasingly interconnected global marketplace[5–7]. In this Letter, we draw on recent theoretical progress in random matrix theory and statistical physics to fundamentally extend and reinterpret May’s theorem. We confirm that a wide range of ecological models become unstable at the point predicted by May, even when the models do not strictly follow his assumptions. Surprisingly, increasing the interaction strength or diversity beyond the May threshold results in a reorganization of the ecosystem – through extinction of a fixed fraction of species – into a new stable state whose properties are well described by purely random interactions. This self-organized state remains stable for arbitrarily large ecosystem and suggests a new interpretation of May’s original conclusions: when interacting complex systems with many components become sufficiently large, they will generically undergo a transition to a “typical” self-organized, stable state.

#11
Ivana Cvijovic
Harvard University
Systems Biology

High-resolution lineage tracking reveals traveling wave of adaptation in laboratory yeast

In rapidly adapting asexual populations, including many microbial pathogens and viruses, numerous mutant lineages often compete simultaneously for dominance within the population. These complex evolutionary dynamics determine the outcomes of adaptation, but they have been difficult to observe directly. While earlier studies used whole-genome sequencing to follow molecular adaptation, these methods have very limited frequency resolution in microbial populations. Here, we introduce a novel renewable barcoding system to observe evolutionary dynamics at high resolution in laboratory budding yeast. We find nested patterns of interference and hitchhiking even at low frequencies. These events are driven by the continuous appearance of new mutations that modify the fates of existing lineages before they reach substantial frequencies. We observe how the distribution of fitness within the population changes over time, finding a “traveling wave” of adaptation that has been predicted by theory. We show that the dynamics of clonal competition create a dynamical rich-get-richer effect: fitness advantages acquired early in evolution drive clonal expansions, which increase the chances of acquiring future mutations. However, less-fit lineages also routinely leapfrog over strains of higher fitness. Our results demonstrate that this combination of factors, which is not accounted for in any existing model of evolutionary dynamics, is critical in determining the rate, predictability, and molecular basis of adaptation.

#12
Margherita De Marzio
Harvard Medical School
Channing Division of Network Medicine

Effect of mechanical compression on the gene expression of Human Bronchial Epithelial Cells

The collective migration and rearrangement of cells in dense epithelial tissues represent the driving force of multiple biological processes, from embryo development to airway remodeling and cancer invasion. The coordinated interplay of mechanical forces and signaling molecular events allows epithelial tissues to adopt different dynamical behaviors: the epithelium sheet typically remains quiescent and non-migratory while performing its routine barrier and immune functions but becomes dynamic and migratory during morphogenesis, repair, invasion and metastasis.
Among the possible feedback mechanisms regulating these dynamical changes, mechanotransduction has been observed to play a key role. Recent experiments have shown that mechanical compression on Human Bronchial Epithelial Cells (HBEC) is able to trigger a transition from a solid-like, “jammed”, cellular state to a fluid, “unjammed”, phase, where cells become highly mobile and exhibit elongated shapes resembling the mesenchymal state of tissues. While several studies have been performed to characterize this transition at the dynamical level, the molecular mechanisms underlying the unjamming response to mechanical stress are poorly defined. In this study, we propose a network approach to identify genome-wide alterations of the genetic expression profile induced by mechanical compression. By combining experimental RNA-Seq data on compressed HBECs and Protein-Protein Interaction (PPI) Network, we use a betweenness centrality measure to determine how the genetic perturbation triggered by pressure propagates through the PPI network. Our results suggest that compression activates a two-step cellular response: an immediate, transient response involving membrane signaling proteins, and a long-time, steady-increase effect that influences genes associated to cell shape, extracellular matrix reorganization and cell motility. More surprisingly, this second cellular response resembles the genetic expression profile of asthmatic HBCEs, where cells typically show a global unjammed behavior.
The dynamical and structural features of HBECs exhibit a tight connection with their genetic signature, shedding light on future experiments to understand such fundamental relation.

#13
Rui Fang
Harvard GSAS
Molecules, Cells and Organisms

Understanding the Functional Dynamics of the 26S Proteasome by Nucleotides-proteasome Interaction Study

The 26S proteasome, consisting of a barrel-shaped proteolytic 20S core complex and one or two 19S regulatory complexes, is a macromolecular machine (2.5 MDa) responsible for regulatory protein degradation in eukaryotic cells. The proteasome has a hexameric ring of AAA+ ATPases at the bottom of the 19S proteasome which converts ATP’s chemical energy to mechanical force to unfold protein substrates and translocate the denatured polypeptide through the central pore into the 20S for degradation. Though genetic, biochemical and structural studies have revealed great details about the proteasome, how the six ATPases in the 19S particle coordinate their ATP cycles to power substrate translocation and conformational transitions of other parts of the proteasome remains elusive. To understand this question, we developed a fluorescent reporter with sub-nanomolar sensitivity to measure different aspects of proteasomal activities. In a competition assay, we found that ATP-γS inhibited substrate degradation much more strongly compared to ADP, and primarily affected the rate of substrate translocation. We then developed a single-molecule assay to study the kinetics of the nucleotides-proteasome interaction. By labeling nucleotides with fluorophores, we are able to detect single nucleotide binding and disassociation events under a TIRF microscope. The result shows that proteasomal ATPases are characterized by distinct nucleotide binding kinetics. Interestingly, the overall binding kinetics of ATP is similar to that of ADP and ATP-γS, while AMP binds rather weakly. To explain the different inhibitory effects of nucleotides in the competition assay, we devised a Markov-state model representing all possible transition of the proteasomal ATPases. Simulation of this model suggests an entropy-driven mechanism underlying the inhibition by nucleotide inhibitors, which may be adaptive under low ATP conditions. Our study provides novel understanding of the working principles of proteasomal ATPases, with the goal to obtain mechanistic insights by bridging the recent structural advances and biochemical observations.

#14
Travis Gibson
Harvard Medical School
Pathology Department

Robust and Scalable Models of Microbiome Dynamics

As the microbiome field moves forward to meaningful clinical applications, there is intense interest in understanding the dynamic behavior of the microbiota, including for purposes of designing bacteriotherapies or "bugs as drugs." Central to the design of such therapeutics is explication of the causal microbial interaction network and population dynamics of the therapeutic organisms as well as underlying microbiota. In this work we present a Bayesian nonparametric model and associated efficient inference algorithm that addresses the key conceptual and practical challenges of learning microbial dynamics from time series microbe abundance data. These challenges include high-dimensional but temporally sparse and non-uniformly sampled data; high measurement noise; and, nonlinear and physically non-negative dynamics. Our contributions include a new type of dynamical systems model for microbial dynamics based on what we term interaction modules, or learned clusters of latent variables with redundant interaction structure (reducing the expected number of interaction coefficients from O(n^2) to O((log n)^2)); a fully Bayesian formulation of the stochastic dynamical systems model that propagates measurement and latent state uncertainty throughout the model; and introduction of a temporally varying auxiliary variable technique to enable efficient inference by relaxing the hard non-negativity constraint on states. We apply our method to a Clostridioides difficile infection gnotobiotic mouse model, and also present preliminary results on a new study with gnotobiotic mice harboring complex microbiota from human donors.

#15
Alexander Golden
Boston University
Physics

Tracing Lineages Through Spatial Transitions

Range expansions in growing populations can have a profound effect on the properties of the lineages of those populations. In some cases, for example in some growing bacterial colonies, range expansion has been found to locally reduce genetic diversity. The mechanisms for understanding how colony growth and morphology can affect population genetics are beginning to be heuristically understood but the connection between the two is still not well understood. In particular, bacterial colonies in two dimensions have been shown to grow into a very diverse family of morphologies by varying only the available nutrient: structures from flat front growth, viscous fingering, dense branching morphologies, to structures resembling diffusion limited aggregation have all been found in the same experimental systems by varying parameters such as nutrient or agar concentration. These dramatic shifts in spatial population structure must certainly have dramatic effects on the population genetics in these colonies, but it is unclear how these transitions are characterized. However, the small number of control parameters necessary to achieve these dramatically different phases suggest that it will be possible to characterize these phases using numerical modeling. We use hybrid PDE-agent based models to investigate how different nutrient and migration properties can affect the population genetics of bacterial colonies growing in 2D. We demonstrate preliminary results suggesting that different levels of nutrient, each giving rise to different colony morphologies, affect the rate of diversity loss.

#16
Anupam Gupta
School of Engineering and Applied Sciences
School of Engineering and Applied Sciences

Motility and adhesion gradient induced vertebrate body axis elongation and somite formation

The body of vertebrate embryos forms by posterior elongation from a terminal growth zone called the Tail Bud (TB). The TB produces highly motile cells that eventually constitute the presomitic mesoderm (PSM), a tissue playing an important role in elongation movements. PSM cells establish an anterior posterior cell motility gradient which parallels a gradient associated with the degradation of a specific cellular signal (Fgf8) known to be implicated in cell motility. As Fgf8 degrades over time, anteriorly positioned cells move less, before eventually coming to a rest as they aggregate into epithelial somites. We show that simple microscopic and macroscopic mechanochemical models for tissue extension that couple Fgf activity, cell motility, cell density and tissue rheology at both the cellular and continuum levels suffice to capture the speed and extent of elongation. These model qualitatively capture the condensation of cells into somites due to the effect of adhesion in the anterior region. These observations explain how the continuous addition of cells that exhibit an increase in cell density and a gradual reduction in motility combined with lateral confinement can be converted into somite formation in the anterior region and an oriented movement and drive body elongation.

#17
Sabina Haque
Harvard University
Systems Biology

Stochasticity and magnetoreception in models of magneto-aerotaxis

Magnetotactic bacteria use external magnetic fields to influence their direction of motility by synthesizing organelles containing magnetic mineral crystals Fe$_3$O$_4$ (magnetite) or Fe$_3$S$_4$ (greigite). These magnetosomes arrange into linear intracellular chains, which imparts a maximal magnetic dipole onto the bacterium and allows it to passively align to external magnetic fields as weak as the geomagnetic field. Several experimental studies have confirmed that magnetotactic bacteria use magnetotaxis in combination with aerotaxis to find optimal concentrations of oxygen. While this process has been theoretically characterized in the literature, there are fundamental limitations to these models. For example, they assume that there is no active biomagnetic sensing happening in the bacterium and ignore stochastic reversals in direction that have been observed experimentally. In addition, while random walk models for chemotaxis in E. coli are common, their usefulness in describing magnetotactic bacteria has been limited. Here, I revise the existing models of magneto-aerotaxis for a single magnetotactic bacteria and investigate if there is still a quantitative advantage (e.g. reduced time taken to find optimal [O2]) when stochastic switching rates and magnetoreception are included. Using simulations and a random walk model, magnetotactic bacteria were confirmed to find optimal oxygen concentrations in less time than non-magnetotactic bacteria.

#18
Matthew Holmes
Harvard University
MCB

Single-molecule imaging reveals distinct subcomplexes of the Bacillus subtilis division machinery

The tubulin homolog FtsZ is integral to bacterial cytokinesis and FtsZ polymers localize in a ring (the Z-ring) at future division sites, where they recruit multiple factors (the divisome) necessary for division. Z-ring constriction and the corresponding synthesis of cell wall between daughter cells are essential for cell division; FtsZ treadmilling is required for both. But what are the specific functions of each component of the divisome, and how do they collectively contribute to cell division? Motions of divisome components were imaged using HaloTag fusions to each protein, labelled with Janelia Fluor dyes. We characterize divisome members as either remaining immobile with FtsZ subunits, or moving around the division site. This latter group contains cell wall transglycosylase FtsW and a subset of non-enzymatic proteins (DivIB, DivIC, and FtsL) which interact with one another but have unknown function. All four proteins move at the division site with velocities comparable to the previously characterized cell wall transpeptidase Pbp2B, and these directional motions require cell wall synthesis. We propose that these proteins move around the cell in complex in a manner dependent on treadmilling FtsZ filaments.

#19
Tanush Jagdish
Harvard University, Harvard Medical School
Systems Biology

Exploring Major Evolutionary Transitions Using Experimental Evolution

Rare events have played an outsized role in shaping the diversity of life on earth. These include the evolution of multicellular division of labor, the eukaryotic transition via the first endosymbiotic event, and the origin of life itself. Over the last few decades, experimental evolution using fast-growing microbes has led to a series of discoveries reshaping our understanding of fundamental evolutionary processes. Yet, as a field, we have largely failed to reproduce rare, major evolutionary transitions within the laboratory. There are three possible reasons for this. First, mechanisms underlying evolutionary transitions might follow the same rules we see in vitro, but are statistically rare events and do not occur in the limited sample size of laboratory evolution. Second, the mechanistic bases of such major transitions might be truly novel, implying that environmental pressures applied in traditional laboratory evolution are unsuited to reproduce them. Third, the genetic architecture of extant model organisms might be constrained by their evolutionary histories, precluding the emergence of complex phenotypes seen in other organisms. These are not mutually exclusive problems, and efforts to surpass them will require creative and novel experimental strategies. Here I will propose and discuss methods to confront each challenge and provide preliminary data from my work with Michael Desai, Andrew Murray, and Jack Szostak to experimentally reconstruct three distinct evolutionary transitions. The goal here is not to provide final answers, but rather to discuss and build an experimental framework aimed at better understanding the rare events that have crucially shaped the diversity of life.

#20
Ofer Kimchi
GSAS
Biophysics

Developing a synthetic post-translational protein oscillator

Synthetically designed protein circuits have the potential to lead to a diverse new class of therapeutics and significantly change the landscape of biomimicry engineering. While synthetic gene circuits have already borne fruit over the past several decades, post-translational circuits engineered to bypass transcription/translation machinery can offer more immediate responses and more direct coupling to endogenous systems. Engineering dynamic phenomena such as oscillations in these circuits remains an outstanding challenge. Only a few known biological systems, such as the KaiABC system regulating the circadian clock in cyanobacteria, offer examples of post-translational oscillators, creating a need for theoretical work to fill in the gaps. We will describe our progress towards developing simple kinetic reaction networks capable of producing oscillations, with a focus on experimental feasibility.

#21
Julian Kimura
Harvard University
Organismic and Evolutionary Biology

Development of Hofstenia miamia and the Embryonic Origin of Neoblasts

Animals that are capable of “whole body” regeneration are able to replace any missing cell type. Species that are able to undergo whole body regeneration often do so using a population of adult stem cells that are effectively pluripotent such as the i-cells of cnidarians and neoblasts of planarians. We seek to identify the embryonic origins of these adult stem cells in the new model system Hofstenia miamia, a highly regenerative acoel species that produces experimentally accessible embryos in the laboratory. Hofstenia possess a population of stem cells, also called neoblasts, that are necessary for regeneration and express homologs of piwi and other markers of planarian neoblasts. We generated a developmental atlas by studying gross morphological changes and cellular movements during Hofstenia embryogenesis. A coordinated cellular movement occurred at about 43-55 hours post laying where the cells on the surface of the animal pole became internalized. Concurrently in development (Dimple stage), bulk RNAseq detected a significant upregulation of genes that mark the neoblasts in the adults. We are now applying single-cell sequencing and embryonic lineage tracing approaches to assess whether neoblasts-like cells emerge at the Dimple stage. The identification and functional characterization of candidate regulators will enable us to determine the mechanisms for neoblast specification during embryogenesis.

#22
Thomas LaBar
Harvard Faculty of Arts and Sciences
Molecular and Cellular Biology

Mutation supply and the genetic mechanisms of cellular evolution

An understanding of which factors alter the genetic mechanisms driving adaptation in different populations is required to predict the outcome of cellular evolution. One potential aspect that may alter these genetic mechanisms is a population’s mutation supply. While large populations with a large mutation supply can access rare, strongly-beneficial mutations, small populations with a small mutation supply can only access common, weaker beneficial mutations. Thus, mutation supply may drive different populations down alternative evolutionary trajectories, especially if potential adaptive mutations are strongly epistatic. I am using a combination of computational modeling and yeast evolution experiments to test whether mutation supply alters the mechanisms underlying adaptation in cellular populations. I will present preliminary results from an evolution experiment with a yeast strain, deficient in transcriptional activation of glycolysis, that has evolved at both a large population size (2x10^8 individuals) and a small population size (6x10^5 individuals). I will also discuss future work on this evolutionary system, specifically concerning the reconstruction of adaptive mutations. The results from this experiment will elucidate how mutation supply alters the mechanisms of adaptation in cellular populations and provide insight into the mechanisms underlying up-regulation of glycolysis, a phenotype relevant to the Warburg effect in cancer.

#23
Anastasia Leshchyk
Worcester Polytechnic Institute
Bioinformatics and Computational Biology

Revealing Tumor Heterogeneity in Triple-negative Breast Cancer at Single-cell Resolution through a De Novo Deep Clustering Approach

The recent emergence of single-cell RNA sequencing (scRNA-Seq) technology has allowed the characterization of heterogeneous cell populations by accurately measuring transcript expression levels in individual cells. This has helped to shed light on the genetic determinants of cellular heterogeneity and to discover new cell types and cell development states. Cellular heterogeneity has been observed in cancers, which could help to explain why some cancers do not respond to treatment and could be targeted for future therapeutics.
Triple-negative breast cancer (TNBC) is known for its high inter- and intra-tumor heterogeneity. TNBC is associated with poor survival prognosis due to its aggressiveness and lack of effective therapies. In this study, we examined primary untreated TNBC tumors from six patients using single-cell RNA-seq data. We implement a novel deep unsupervised single-cell clustering method (DUSC) to learn an unbiased, informative and robust representation of the scRNA-Seq data.
DUSC helps to reveal new subpopulations of cells sharing common transcriptomic features and providing new biological insights to the TNBC heterogeneity. Our analysis reveals subgroups of cells shared among all patients and gene signatures associated with the poor and good patient survival rates. We propose DUSC as a new method to identify tumor subclones accentuated by copy-number variation, with the copy numbers strongly correlating with the patient survival prognosis.

#24
Ethan Levien
Harvard SEAS
Applied Math

Quantifying phenotypic variability in finite populations

The prevalence phenotypic variability within genetically identical populations is well established, yet the mechanisms underlying this variability remain poorly understood, as do the consequences for population dynamics. Answering these questions requires a mathematical theory relating data obtained in microfluidic devices, where the variability and heritability of phenotypic traits can be observed, to measurements made in competition experiments where fitness can be quantified. Previous studies have shown how to relate the statistics of phenotypes throughout an exponentially proliferating population to the population growth. However, the lineage statistics obtained in a finite population are not equivalent to those in an exponentially proliferating population due to the non-ergodicity of the growth process. We analyze two methods for estimating the fitness of a population based on data obtained in a finite population — one is based on the dilution rate of the culture while the other uses the distribution of phenotypes. We then derive a relationship between these two estimates in terms of the strength of environmental fluctuations, allowing us to determine whether heritability of phenotypic traits is due to intrinsic or environmental factors. We then apply our theory to data obtained by tracking E. Coli in a microfluidics device and reveal hidden environmental variability. Finally, we propose future experiments that could leverage our results to elucidate the interplay between environmental and intrinsic variability in microbes.

#25
Jie Lin
SEAS
Applied physics

Coevolution of multiple growth traits in microbial populations under serial dilution

The relative fitness of mutants in a microbial population depends on multiple cellular traits. In the most widely-used evolution experiment protocol, serial dilution (where cells grow, enter stationary phase, and are diluted into a fresh medium), three major traits determining fitness are the population growth rate, lag time (the duration of time cells do not grow after diluted into a fresh medium), and yield (number of cells per unit resource). Here we investigate how these traits coevolve in laboratory evolution experiments using a minimal model of population dynamics, where the only interaction between cells is competition for a single limiting resource. We find that the fixation probability of a beneficial mutation depends on a linear combination of its growth rate and lag time relative to the background strain. The relative selective pressure on growth rate and lag time is set by the dilution factor; for example, a larger dilution factor favors the adaptation of growth rate over the adaptation of lag time. This result applies equally to the regime of large populations and high mutation rate, where there is abundant clonal interference, as well as the regime of rare, sequential mutations. Moreover, we show that an emergent correlation between growth rate and lag time at the population level can evolve even if mutations have uncorrelated effects on the two traits.

#26
Nathan Lord
Harvard
Molecular and Cellular Biology

Co-receptor feedback shapes the Nodal signaling gradient

Embryos must communicate instructions to their constituent cells over long distances. These instructions are often encoded in the concentration of diffusible signals called morphogens. In the textbook view, morphogen molecules diffuse from a localized source to form a concentration gradient, and target cells infer their positions in the embryo by measuring the local morphogen concentration. However, natural patterning systems often incorporate extensive feedback on signaling, suggesting that reliable patterning requires precise control. For example, the mesendoderm inducer Nodal drives production of diffusible inhibitors of its own signaling, as well as of components of its cell surface receptor complex. We previously showed that negative feedback allows the embryo to correct perturbations to the Nodal signaling gradient. However, the role of Nodal-driven receptor expression in patterning remains unclear. Here, we examine how feedback regulation of Oep, a co-receptor required for Nodal signaling, shapes the Nodal gradient in early zebrafish embryos. We show that Oep constrains Nodal ligand spread: signaling ranges of the Nodal signals Cyclops and Squint are markedly expanded in maternal-zygotic oep mutants. Despite being regarded as a permissive co-factor, we find that increasing Oep levels sensitizes cells to Nodal ligands. Finally, we show that zygotic oep expression is required to maintain sensitivity to Nodal ligands and to restrict ligand spread. We suggest that Nodal-driven expression of oep creates a feedback loop that regulates Nodal ligand mobility and potency

#27
Victor Luria
HMS
Systems Biology

Variation and novelty in evolution: novel genes continuously arise, enable protein structural innovation and function in the brain

How new protein-coding genes originate is a central, unsolved evolutionary question. Most genes were thought to arise by copying or transferring existing genes. Long thought impossible to arise from non-coding sequence, novel genes that arise de novo from genomic "junk" DNA or from long non-coding RNA have recently found in eukaryotic genomes. Novel genes are taxon-restricted, being present in one or few species, and may encode structurally new proteins. Strikingly, novel genes are invariably expressed in the brain and germline. We initially found a taxon-restricted gene, APCDD1, and showed it functions in neurons and skin in humans and other chordates. To understand how novel genes appear, what proteins they make, what functions they have and what their general properties are, we combined mathematical, computational and experimental approaches. To evaluate how often may novel genes arise, we built a mathematical model based on gene and genome parameters and dynamic factors such as mutation. We found genomes should make many new genes and keep few. We computationally identified candidate novel genes in 25 eukaryotic genomes using phylostratigraphy and proteomics data and evaluated their predicted biophysical properties. Compared to ancient proteins, novel genes encode proteins that are shorter, fragile, disordered, promiscuous yet less prone to forming toxic prions or to aggregation. Third, we biophysically compared novel proteins to ancient proteins, we tested novel gene function in vivo in zebrafish brains using CRISPR inactivation, and we showed candidate novel genes are expressed in the human brain at multiple ages. We found that genomic sequence is turned over such that many novel genes arise continuously and encode short proteins, of which a small fraction perdures evolutionarily. The survivors encode proteins with distinct structural features and are expressed in the brain, suggesting genomes continually generate variation that enables new structures and functions, and is selected upon.

#28
Anastasia Lyulina
Harvard Medical School
Systems Biology

Population genomics of the critically endangered spoon-billed sandpiper

What makes a species vulnerable to extinction? We study the population genomics of the critically endangered spoon-billed sandpiper and its sister species, the red-necked stint, which is of least concern. Using whole-genome data, we found that while the red-necked stint population was relatively constant, the spoon-billed sandpiper population peaked during the last glacial maximum and has been declining since. Whereas low genetic diversity is thought to be among the main causes of extinction, we identify a different risk to the spoon-billed sandpiper. Our simulations show that the population growth that occurred prior to a bottleneck led to an overabundance of segregating deleterious alleles, imposing an additional burden on the spoon-billed sandpiper population by increasing the cost of inbreeding.

Reference: Mateusz Konczal et al., in preparation.

#29
Elizabeth May
GSAS
Molecules, Cells and Organisms

Molecular and Cellular Mechanisms of Neuronal Connectivity

It is increasingly clear that many neurological disorders manifest physiologically as a miswiring of neural circuits in the brain – from an individual neuron’s perspective, a case of mistaken identity when it comes to finding and making appropriate connections. I investigate how neurons discriminate between correct and incorrect cell-cell contacts to determine their synaptic partners. Neuronal proteins called clustered protocadherins are important markers of cell identity and regulators of neuronal connectivity; juxtaposed cells in the brain compare their unique sets of cell-surface protocadherins to distinguish one another. Several genetic studies have identified mutations in clustered protocadherin genes that correlate with increased incidence of autism spectrum disorder, schizophrenia and bipolar disorder. In my poster, I present my exploration of one specific question to understand how clustered protocadherins signal to the neuron to either synapse with or avoid another cell. What do the protein-protein interactions look like when two cells compare their respective sets of clustered protocadherins? I use a combination of X-ray crystallography for structure determination and functional assays to test the interaction properties of clustered protocadherins. Deciphering the molecular interactions of clustered protocadherins is key to understanding how they function as cell identity markers to facilitate neural circuit wiring in the brain.

#30
Harry McNamara
GSAS
Physics

Electrical Reaction-Diffusion Pattern Formation in Synthetic Biological Tissues

Electrical signaling in biology is typically associated with action potentials, transient spikes in membrane voltage that return to baseline. More generally, electrical tissues are reaction-diffusion systems which can support patterns which are structured in space but stationary in time. It is possible that bioelectrical signaling could coordinate biological processes on timescales which are slower than action potentials – for example, during embryonic development. Constraints on electrophysiological measurement in vivo have made it challenging to assess these hypotheses quantitatively.

By engineering electrically inert mammalian cells to express particular ion channels along with optogenetic tools to mediate all-optical electrophysiology, we are building bioelectric tissues from the bottom-up to investigate fundamental aspects of biological pattern formation. In one demonstration, we engineer demonstration, we engineer spiking cells which support traveling action potential waves, spiral waves, and transitions to chaotic dynamics. Using patterned illumination, we study these tissues as an excitable medium: for example, we show that wavefront curvature influences conduction velocity, and that dynamical transitions to instability and chaos depend on the geometry of spatial coupling in the tissues. We also show these cells can be wired into cellular circuits capable of simple information processing and memory (e.g. oscillators).

Finally, we engineer an electrically bistable cell line, and show theoretically and experimentally that homogeneous or nearly homogeneous tissues of bistable cells can undergo spontaneous symmetry breaking into electrical domains with different resting potentials, separated by bioelectrical domain walls. We map bioelectrical domain wall motion using all-optical electrophysiology in an engineered cell line and in human iPSC-derived myoblasts. These results demonstrate a novel form of bioelectrical pattern formation and long-range signaling that may coordinate differentiation of myoblast precursors into mature muscle tissue.

#31
Matthew Melissa
GSAS
Physics

Traveling-Wave Models of Evolution: The Generalized Infinitesimal Limit

The evolutionary dynamics of large populations are rather complicated. A large number of distinct adaptive lineages may be present in a population at once, such that the fate of a new mutation cannot be considered in isolation. The dynamics are well-understood in several limiting cases. These can largely be partitioned into cases in which mutations are strong—that is, in which they fix with probabilities differing substantially from that of a neutral mutation—and cases in which mutations are weak, such that they are only subject to selection collectively. Here we extend these approaches to consider the dynamics of populations subject to mutations which are neither entirely strong nor entirely weak. We discuss how quantities such as the overall rate of adaptation and the coalescence timescale depend on the range of potential effects that mutations can confer and other population-genetic parameters. We further identify a few key timescales and fitness scales—the time required for the fittest lineages to deterministically sweep through the population, the range of fitnesses which routinely contribute future common ancestors of the population, and the most-likely effect size of a fixed mutation. We discuss the suitability of different approximation schemes in terms of these scales, and argue that many dynamical quantities of interest depend on these scales in a rather universal way.

#32
Yuchuan Miao
Harvard Medical School
Genetics

Feedback control of the human segmentation clock

The metameric patterning of the human body axis is established early in embryogenesis by segmentation of the paraxial mesoderm. This process involves the sequential formation of somites, which give rise to vital structures such as vertebrae and skeletal muscles. Despite the major developmental importance, human segmentation is poorly understood owning to the difficulty to access early stage embryos.

In model organisms, segmentation is controlled by a molecular oscillator known as the segmentation clock, where Wnt, FGF and Notch pathways and their regulated gene expression are rhythmically activated. This oscillator has been shown to be an excitable system, with Yap mediated mechanical cues modulating its threshold for activation. However, the molecular crosstalk schemes, detailed regulation mechanisms, and biological functions of excitability remain unknown.

Recently, human induced pluripotent stem (iPS) cells have been successfully differentiated into displaying oscillations, demonstrating the existence of a human segmentation clock. Further, these human oscillators displayed conserved molecular characteristics and excitable properties. Here, guided by the broad theoretical framework of excitability, I propose to use the iPS cells derived in vitro system to elucidate the mechanisms of human segmentation. Specifically, the proposed experiments address the molecular and cellular feedback loops, the regulatory schemes by mechanical information, and the essential functions of the excitable segmentation clock. I will combine various molecular and cell biological approaches, including CRISPR-Cas9, optogenetics, single cell RNA sequencing, live cell imaging, etc. Together, these efforts will illuminate general principles of self-organization in human development and contribute to treating diseases caused by dysregulated segmentation.

#33
Danai Montalvan
SEAS
Applied Physics

A synthetic nucleus for synthetic cells

Synthetic cells can currently be produced by different methods, most of which include encapsulating cellular extracts, enzymes, or structural proteins. However, even in these simplified cells, precise control of the reactions occurring inside is still difficult. A solution to this problem is to control the reactions with light. Inspiration for this approach comes from optogenetics, where light-exposure can induce gene transcription, protein synthesis, and protein folding. As a first step toward this goal, we have designed synthetic cells in which light can control DNA replication. The “genome” of the synthetic cell includes template DNA strands grafted onto colloidal particle. When illuminated with a specific wavelength of light, the system heats up. We use the light-driven heating to melt DNA duplexes and control whether DNA polymerases can attach to the template. Specifically, we design our sequences to using the NUPACK software. The genome sequence is prone to self-priming, as it contains a pair of complementary sequences separated by a hairpin. Both a hairpin and a blocking-oligo control the ability of the genome to self-prime. Our predictions show that low temperatures favor the genome/blocking-oligo interaction while higher temperatures favor genome self-priming and, therefore, DNA replication in the presence of polymerases. We observe the DNA production using a duplex-intercalating fluorophore. I will discuss how temperature and light affect DNA production in the synthetic nucleus. Also, I will address the feasibility of making synthetic cells by encapsulating clusters of DNA-grafted colloids in porous microparticles.

#34
Emily Moore
Harvard School of Dental Medicine
Developmental Biology

The role of Galphas signaling in postnatal tooth development

Dental stem cell (DSC) differentiation is perhaps regulated by Wnt and Hedgehog (Hh) signaling pathways, but these mechanisms are poorly understood. Axin2 and Gli1 are components of the Wnt and Hh pathways, respectively, and can be used to identify potential DSCs. Our lab recently demonstrated that a G-protein subunit (Gas) may regulate the balance of Wnt/Hh signaling to facilitate stem cell differentiation. We therefore aim to identify potential Wnt/Hh-responsive DSCs and determine whether Gas balances these pathways to direct development. We generated mouse models to simultaneously track and eliminate or activate Gas signaling in Wnt/Hh-responsive cells in vivo. tdTomato (tMt) production and the Gas mutation was induced in pups from postnatal day 7-14, which were then sacrificed at 4-6 weeks for histology and FACS/ RNAseq. tMt+ cells were located in the dental pulp, dentin, incisor mesenchyme/epithelium, incisor alveoulus, mandible, and alveolar bone. Groups with activated Gas signaling exhibited thicker molar dentin and fibrous tissue formation instead of alveolar bone beneath the molar root. Gli1 groups lacking Gas signaling had stunted mandible growth, while the Axin2 groups demonstrated shortened incisors. Mice containing activated and disrupted Gas signaling had significantly more labeled cells in all examined tissues. RNAseq revealed that targets in the Wnt, Hippo, cAMP, and calcium signaling pathways were altered, as well as targets involved in stem cell pluripotency, osteogenic differentiation, and osteoclast activation. Our data suggest that Wnt/Hh-responsive cells are present throughout the jaw and give rise to tooth-forming progeny. More importantly, our results indicate that Gas signaling directs normal tooth formation. We also identified alternative signaling pathways and specific target genes that are involved in Gas-mediated Wnt/Hh signaling in the incisor. Overall, we have made strides to distinguish potential DSCs and related signaling pathways that can be therapeutically targeted to generate bone for tooth regeneration.

#35
Maria Mukhina
Harvard
MCB

Towards mechanoluminescent force nanosensor for biological systems

Mechanical force patterns underlie the functionality of many biological systems. Exploration from this perspective is blocked by lack of a suitable in vivo force detection tool. I am working to develop such a tool, using mechanoluminescent (ML) nanocrystals as non-invasive stress sensors for living cells. My project is inspired by emerging evidence that global force patterns, as actual mechanical waves, orchestrate basic genomic processes. Defects in such patterns confer genome instability, birth defects and cancer. I have identified the mechanism of mechanoluminescence in ZnS:Mn microparticles and showed that it involves the synergistic interaction between pressure-initiated piezoelectric polarization, complex internal electric structure and trapped charges within the faulted ZnS:Mn polycrystal. The effect is localized to nanoscale spatial range and the threshold for ML appearance in ZnS:Mn is only 200 kPa what makes it perfect candidate for nanosensor of intranuclear forces. I have designed ML nanocrystals that will emit fluorescent light in response to appropriately low forces and can be readily targeted to chromosomes.

#36
Anjalika Nande
Harvard University
Physics and Program for Evolutionary Dynamics

The Role of Drug Kinetics on the Evolution of Resistance.

Emergence of drug resistance due to treatment non-adherence is a problem especially in chronic prolonged viral infections like the Human Immunodeficiency virus (HIV) and Hepatitis B (HBV) and C (HCV) viruses. Long acting drugs are being developed as one way to address this problem. Though this promises to be useful in the context of treatment adherence, we do not yet know how this would affect resistance. With this in mind, we analyze the effect of dosing intervals on the establishment of resistance due to mutants existing prior to treatment and those that are produced during treatment in the presence of time-dependent drug profiles. We find that there exists an initial time-frame after treatment initiation that has the most influence on this establishment probability. Depending upon the nature of the drug kinetics during this time as well as infection parameters, increasing the dosing interval might be better or worse for the establishment of resistance. Our results suggest that drug kinetics affect selection and competition in the system in a complicated manner and should be factored in while designing new treatment strategies.

#37
Maximilian Nguyen
Harvard Medical School
Systems Biology

Reciprocity: A Functional Measure of Coupled Transcription Factor Binding

The mode of interaction of transcription factors (TFs) on eukaryotic genomes remains a matter of debate. Single-molecule data in living cells for the TFs Sox2 and Oct4 were previously interpreted as evidence of ordered assembly on DNA. However, the quantity that was calculated does not determine binding order but, rather, energy expenditure away from thermodynamic equilibrium. Here, we undertake a rigorous biophysical analysis which leads to the concept of reciprocity. The single-molecule data imply that Sox2 and Oct4 exhibit negative reciprocity, with expression of Sox2 increasing Oct4’s genomic binding but expression of Oct4 decreasing Sox2’s binding. Models show that negative reciprocity can arise either from energy expenditure or from a mixture of positive and negative cooperativity at distinct genomic loci. Both possibilities imply unexpected complexity in how TFs interact on DNA, for which single-molecule methods provide novel detection capabilities.

#38
Alan Pacheco
Boston University
Bioinformatics

Generation of microbial community phenotypes through reinforcement learning

Environmental composition is a key factor in defining the structure and function of microbial ecosystems. This environment-phenotype axis has been the subject of many computational and experimental studies, which have revealed detailed molecular mechanisms that drive these relationships in some small microbial consortia. Despite this progress, it remains difficult to scale this detailed mapping between environment and phenotype to large, complex communities. This is because natural microbial communities can harbor up to thousands of different organisms and environmental substrates, making up a vast combinatorial space that precludes exhaustive experimental testing and computational prediction. Here, we present a method based on reinforcement learning and dynamic flux balance analysis (dFBA) that selects optimal environmental compositions to produce target community phenotypes. In this framework, dFBA is used to model the growth of a community in candidate environments. Reinforcement learning is then used to evaluate the behavior of the community relative to a target phenotype, and subsequently adjust the environment to allow the organisms to more closely approach this target. We apply this iterative process to model communities of varying sizes, showing how it can rapidly identify environments that yield desired phenotypes, saving time and computational resources. Moreover, this combined approach produces testable predictions for the in vivo assembly of microbial communities with specific properties, and can facilitate rational environmental design processes for complex microbiomes.

#39
Vinuselvi Parisutham
UMASS Medical School Worcester
Programs in Systems Biology

Influence of Chromosomal Position in Controlling Prokaryotic Gene Expression

In Prokaryotes, the location of a gene on the chromosome is important for maintaining correct gene expression levels and gene translocations can strongly alter organism fitness. While gene dosage plays a role in this phenomenon, even translocations that maintain gene dosage often have a deleterious effect. What is less clear is how the quantitative response of a gene to a particular concentration of TFs depends on gene location. As such, we measure how regulation by transcription factors (TFs) is altered by the chromosomal location of the target gene. To interpret this data, we invoke a thermodynamic model of gene regulation which has been extensively used to predict regulation levels as a function of the number of TFs, affinity of TF binding and number of targets genes. We find that regulation of the target gene aligns with the predictions of the thermodynamic model with an “affinity” that varies depending on chromosomal location. We interpret the cause for this change in affinity and investigate how it depends on binding sequence and TF identity. These differences in affinity translate to important differences in not only steady-state responses but also the dynamic time-ordered response of the gene. As such, a detailed characterization of the impact of chromosomal position on gene regulation is a crucial step towards generating an accurate predictive model of gene expression and regulatory patterns.

#40
Philip Pearce
Massachusetts Institute of Technology
Mathematics

Flow-induced symmetry breaking in growing bacterial biofilms

Bacterial biofilms are matrix-bound multicellular communities. Biofilms represent a major form of microbial life on Earth and serve as a model active nematic system, in which activity results from growth of the rod-shaped bacterial cells. In their natural environments, from human organs to industrial pipelines, biofilms have evolved to grow robustly under significant fluid shear. Despite intense practical and theoretical interest, it is unclear how strong fluid flow alters the local and global architectures of biofilms. Here, we combine highly time-resolved single-cell live imaging with 3D multi-scale modeling to investigate the effects of flow on the dynamics of all individual cells in growing biofilms. Our experiments and cell-based simulations reveal that, in the initial stages of development, the flow induces a downstream gradient in cell orientation, causing asymmetrical droplet-like biofilm shapes. In the later stages, when the majority of cells are sheltered from the flow by the surrounding extracellular matrix, buckling-induced cell verticalization in the biofilm core restores radially symmetric biofilm growth, in agreement with predictions from a 3D continuum model.

#41
Rostam Razban
Harvard University
Chemistry & Chemical Biology

Protein melting temperature and protein folding free energy are not interchangeable in light of protein evolution

The protein misfolding avoidance hypothesis (MAH) explains the universal negative correlation between protein abundance and sequence evolutionary rate across the proteome by identifying protein folding free energy (ΔG) as the confounding variable. Abundant proteins resist toxic misfolding events by being more stable, and more stable proteins evolve slower because their mutations are more destabilizing. Direct supporting evidence consists only of computer simulations. A study taking advantage of a recent experimental breakthrough in measuring protein stability proteome-wide through melting temperature (Tm) (Leuenberger et al. 2017), found weak MAH support for the Escherichia coli proteome, and no support for the Saccharomyces cerevisiae, Homo sapiens and Thermus thermophilus proteomes (Plata and Vitkup 2017). I find that the nontrivial relationship between Tm and ΔG and inaccuracy in Tm measurements by Leuenberger et al. 2017 can be responsible for not observing strong positive abundance–Tm and strong negative Tm–evolutionary rate correlations.

#42
Annika Rohl
Harvard Medical School
Mathematics

PageRank on Multilayer Networks

With the modern profiling technologies for diverse “-omics" we are able to obtain different types of “-omic” data-sets, for specific diseases or phenotypes. Each datatype can be analyzed independently, using a variety of different methods leading to a better understanding of biological processes. While a single profiling of one molecular type provides limited information, the analysis of the combined "-omics” data gives insight to complex interactions between diverse types of molecules.
To incorporate several “-omic” datatypes, different approaches exist, where we will focus here on network analysis. For example, Gene Regulatory Networks (GRNs), Protein-Protein Interaction Networks (PPIs), or metabolic networks are analyzed, e.g. to study complex diseases.

However, these networks are again connected to each other (proteins can be transcription factors for genes or catalyze metabolic reactions) and mathematically speaking we deal with multi-layer networks (MLNWs).

There are different methods which were successfully applied on single layer networks. One of them is based on the PageRank algorithm (the underlying method of Google’s search engine). It was used for prioritizing disease genes or identifying genes responsible for adverse drug reactions.

There exist a method extending the PageRank algorithm to MLNWs, where each layer consist of the same set of nodes. We applied this method onto a GRN and a PPI, including edges connecting the two. Doing so, we get a ranking of proteins integrating information obtained by the GRN. Our next step is to analyze the impact of the high-ranking genes on the robustness of the MLNW.

However, integrating different “-omic” datatypes lead to MLNW, where the nodes in different layers may differ. Therefore, we started to develop a PageRank-algorithm, where the layer can contain different nodes. Using this algorithm, we can include a metabolic network, such that we can detect the effect of a perturbed gene onto the metabolic layer.

#43
John Russell
SEAS
Applied Physics

Phenotypic Tracking of Microbes with Raman Spectroscopy

Microbial cells have evolved a vast array of mechanisms to adapt to different environments. Many of these adaptations are carried out by changing gene expression levels that can be transmitted between generations and acted upon by natural selection. While much is known about the molecular basis of these changes in single cells, it is difficult to investigate large-scale change in gene expression or cell state, since most global assays are terminal or rely on whole-population measurement.

Recent work (Kobayashi-Kirschvink et al. 2018) has described the use of Raman spectroscopy, which provides a readout of the chemical composition of a sample by probing the vibrational modes of molecules, for classification of microbial cells that had been exposed to different stress conditions. This work has shown that 1. Raman spectroscopy is a viable way of measuring of cells’ phenotypic states and 2. there is a straightforward mapping between Raman spectra and the gene expression levels as measured by RNAseq.

I am working on developing Raman spectroscopy into a technique for monitoring the internal states of single cells through time. I am pursuing technical developments to allow for higher throughput tracking of cells and for live cell measurements, in contrast to previous work with fixed cells. I am also working to quantitatively model these dynamics. Drawing from established work in stochastic processes I plan to create a probabilistic description of these cellular changes. Such a model, when applied to single cells, would allow me to capture both deterministic trends and stochastic cell to cell variation, both of which may be important to understanding the evolutionary significance of these dynamics.

#44
Carlos Sanchez
Harvard Medical School
Systems Biology

Ultraprecise Measurement of Fitness in Bacteria

Quantifying fitness improvement is of key interest to understand adaptation dynamics in evolutionary biology, yet the state-of-the art methods for relative fitness measurement are several orders of magnitude far from the necessary resolution. Here, we describe a microfluidics based competition assay that achieves a fitness resolution down to 10-4 between different strains, without any reporter burden or risk of mutational take-over, unlike their bulk equivalents. We demonstrate the potential of this assay by determining fitness burden resulting from differential expression of proteins from a promoter library.

#45
Georgia Squyres
Harvard University
MCO

Z ring assembly is regulated by FtsZ filament binding proteins

Cell division in bacteria is orchestrated by a group of proteins that work together to carry out cytokinesis and synthesize new cell wall at the division site. Filaments of FtsZ, a bacterial homolog of tubulin, form a “Z ring” at the middle of the cell that constricts as the cell divides, through a mechanism which remains poorly understood. Recently, we have shown that FtsZ filaments treadmill around the division site, and that these dynamics are of key significance in bacterial cell division.
Now, we investigate how FtsZ filament assembly and treadmilling dynamics are regulated. Of particular interest are the FtsZ binding proteins, which are known to bind directly to FtsZ at the division site, and which have been proposed to regulate both FtsZ dynamics and bundling in vitro. We ask whether and how FtsZ binding proteins control filament structure and/or dynamics during the bacterial cell cycle. To investigate this, we use live-cell imaging to characterize the morphology of individual FtsZ filaments and of the Z ring, and to measure the lifetimes of single FtsZ monomers as a precise quantitative reporter of FtsZ’s treadmilling dynamics. Surprisingly, we find that the FtsZ binding proteins do not regulate FtsZ filament kinetics directly, but rather mediate Z ring assembly through filament bundling. Functional regulation of filaments by bundling is a common feature of eukaryotic cytoskeletal systems, but has rarely been seen in bacteria. We propose that this regulated FtsZ filament bundling is a prerequisite for normal Z ring formation and cytokinesis.

#46
Elad Stolovicki
SEAS
Physics

Drop chemostats on a chip

The adaptation process of biological system to novel challenge has a lot of variability. To investigate the spectrum of possible adaptation trajectories requires large ensemble of identical twin population. To address this challenge, we are developing a drop-based microfluidic device with hundreds of chemostats on a single chip. The chemostat is a continuous-culture apparatus that enables growth of cells in a well-controlled environment. The controlled conditions of the chemostat enable the measurement of population response to specific factor by varying only one environmental factor at a time. In the drop microfluidic approach, the chemostat vessel is ~1µL media drop surrounded by inert oil. Every population in every drop is an independent chemostat population. Each chemostat droplet is continuously diluted with fresh media. The flow of the chemostat drop insure the mixing of nutrient and suspension of the cells. The surrounding oil reduce the fouling of cells to the channel walls. The measurements can be done on the whole chemostat drop or only on the subtracted fraction. The subtracted fraction from the chemostat can be used to monitor the cells and environment in the drop using variety of distractive assays without interfering the chemostat experiment.

#47
Charlotte Strandkvist
Harvard Medical School
Systems Biology

Fluctuation-response relations in biological systems

In both natural and synthetic biological systems, we are interested in the response behavior of cells. Specifically we are interested in the ability of cells to respond sharply to specific signals while operating reliably in a wide range of environmental conditions. We present analytical results showing a fundamental trade-off between the sensitivity, or the sharpness of the cellular response, and the cellular noise for a general class of systems in non-equilibrium steady state.

#48
Lei Sun
Harvard Medical School
Systems Biology

Stochastic gene expression influences the selection of antibiotic resistance mutations

Bacteria can resist antibiotics by expressing enzymes that remove or deactivate drug molecules. Here, we study the effects of gene expression stochasticity on efflux and enzymatic resistance. We construct an agent-based model that stochastically simulates multiple biochemical processes in the cell and we observe the growth and survival dynamics of the cell population. Resistance-enhancing mutations are introduced by varying parameters that control the enzyme expression or ecacy. We find that stochastic gene expression can cause complex dynamics in terms of survival and extinction for these mutants. Regulatory mutations, which augment the frequency and duration of resistance gene transcription, can provide limited resistance by increasing mean expression. Structural mutations, which modify the enzyme or efflux efficacy, provide most resistance by improving the binding affinity of the resistance protein to the antibiotic; increasing the enzyme’s catalytic rate alone does not contribute to resistance. Overall, we identify conditions where regulatory mutations are selected over structural mutations, and vice versa. Our findings show that stochastic gene expression is a key factor underlying efflux and enzymatic resistances and should be taken into consideration in future antibiotic research

#49
Tony Tsai
Harvard Medical School
Systems Biology

An adhesion code enables robust pattern formation in the zebrafish spinal cord

An outstanding question in embryonic development is how different cell types reach their final positions correctly, despite large scale cellular re-arrangement during tissue morphogenesis. To achieve this, cell fate specification needs to be coordinated with regulated cell migration and adhesion, yet the mechanisms are not well-understood. In the zebrafish spinal cord, thirteen neural progenitor domains, each consisting of distinct cell type, are specified along the ventral-to-dorsal axis by opposing gradients of Shh and BMP/Wnt. The Shh signal is noisy, resulting in cells specified in a mixed pattern, requiring that individual cells be sorted to resolve this mixed pattern into well-separated domains.
Here we set out to test if differential adhesion plays a role in assisting pattern formation during zebrafish spinal cord development. We developed two ex vivo mechanical assays to measure adhesion forces and adhesion preferences among three types of neural progenitors (p3, pMN, and p0 cells). Interestingly, each cell type exhibited preference to stabilize homotypic contact and adhered more strongly to cells of the same type. A subsequent transcriptomic and genetic analysis revealed three adhesion molecules (N-cadherin, Cadherin 11, and Protocadherin 19) that are differentially expressed among the three cell types. When the expression levels of these adhesion molecules are perturbed, the adhesion preference to cells of the same type is lost, and the neural progenitor pattern in the spinal cord is disrupted in vivo. These findings allow us to propose an “adhesion code” mechanism that promotes cell sorting during tissue scale patterning. While the conventional view of spinal cord patterning is heavily focusing on interpretation of biochemical signals and transcriptional regulation, our findings suggest cell adhesion is a critical contributor to enable precise patterning in a tissue that is undergoing significant morphogenesis.

#50
James Valcourt
GSAS
Systems Biology

Understanding lineage commitment and context-dependent response to TGF-B superfamily signals during early mammalian development

During mammalian development, progenitor cells differentiate and commit to ever-more-specific lineages. The cell’s response to signaling molecules changes with commitment: selecting a given lineage means reinterpreting signals that would have previously promoted an alternative lineage. This process of committing to a lineage by changing response to signaling molecules is of critical importance for proper development, but the dynamics and mechanism that govern the process are unclear. We investigate the changes in human embryonic stem cell response to BMP and Activin A during commitment to the bipotent ectoderm lineage by discovering a two-gene reaction coordinate that allows us to monitor the dynamics of differentiation in real time. We show that knowledge of the expression levels of OCT4:RFP and SOX2:YFP in a double-tagged reporter line is sufficient to predict the cell’s response to BMP4 and Activin A; in contrast, classical neural/ectodermal marker genes activate too late to be useful for this purpose. From a single differentiating population, we isolate cells that are pre- and post-commitment to the neural/ectodermal lineage based on their OCT4:RFP and SOX2:YFP levels, and we analyze these two populations in detail using RNA-seq, ATAC-seq, and ChIP-seq. Using these data, we uncover the gene regulatory network that underlies bipotent ectoderm lineage commitment, and we make specific perturbations to this network to speed or slow commitment. Interestingly, chromatin accessibility at pluripotency-related loci is not associated with commitment to the bipotent ectoderm lineage. These results provide high-time-resolution details of the commitment process and suggest that some of the changes associated with differentiation may be dispensable for initial lineage commitment.

#51
Debra Van Egeren
Harvard University
Systems Biology

Recording transcriptional and cell division histories in individual hematopoietic progenitor cells

Hematopoietic stem cells differentiate through intermediate progenitor states to produce all mature blood lineages. While some restricted intermediate progenitors have been identified and characterized, many parts of the differentiation hierarchy are still unclear. Furthermore, the in vivo kinetics of hematopoietic differentiation have not been fully explored on a single cell level, so the variability in single cell developmental trajectories remains mostly unknown. Finally, the relationship between differentiation and cell division in hematopoietic differentiation is still under debate. To study these questions, we are developing an experimental system which can measure both current transcriptional cell state as well as how long it has been since the cell has undergone a particular differentiation event.

#52
Suyang Wan
HAVARD
Systems Biology

Genome-scale Time-course Genotype-Phenotype Mapping with Promoter-based Barcoding System

Phenotype-genotype mapping is fundamental in
biology. Here we propose a genome-scale single
cell level phenotype-genotype mapping over multiple
generations with high-throughput mother
machine imaging and a barcoding system based
on expression dynamics driven by different promoters.

#53
Shou-Wen Wang
Harvard Medical School
Systems biology

Emergence of collective oscillations in adaptive cells

In multicellular development, oscillatory signalling through a process known as dynamical quorum sensing is often employed to orchestrate group behaviour. Although cell-to-cell communication has been recognised as essential to enable the collective behaviour, how individual cells come out of quiescence and oscillate synchronously with their peers varies from system to system.
Here, we report a generic link between adaptive response and auto-induced collective oscillations. By considering the response of cells to the extracellular signal and vice versa, we establish a general condition for the threshold cell density and the onset frequency of oscillations. The condition in particular requires cells to be phase-leading against the signal, which is possible only when active intracellular processes participate. We present a mathematical proof that adaptation is such a process where the anticipatory phase behaviour occurs over a range of frequencies. The adapt-to-oscillate scenario is shown to be present in several known examples of dynamical quorum sensing as well as in systems which may be considered pathological with regard to the underlying biology.

#54
Xin Wang
Harvard Medical School
None, I am a postdoc

Overcome Competitive Exclusion in Ecosystems

Explaining biodiversity in nature is a fundamental problem in ecology.
An outstanding challenge is embodied in the so-called Competitive
Exclusion Principle: two species competing for one limiting resource
cannot coexist at constant population densities, or more generally,
the number of consumer species in steady coexistence cannot exceed
that of resources. The fact that competitive exclusion is rarely
observed in natural ecosystems has not been fully understood. Here we
show that by forming chasing triplets among the consumers and
resources in the consumption process, the Competitive Exclusion
Principle can be naturally violated. The modeling framework developed here is broadly applicable and can be used to explain the biodiversity of many consumer-resource ecosystems and hence deepen our understanding of biodiversity in nature.

#55
Sean Wilson
Harvard University
Molecules, Cells and Organisms

The Role and Regulation of Cell Wall Hydrolases in Bacillus subtilis

Bacteria are encased in a rigid meshwork called the cell wall. In order to grow, bacteria must continuously remodel their cell wall, inserting new material and breaking old bonds. Recent research has explained many of the mechanisms of cell wall insertion, but the regulation of bond breakage is still not understood. The cell wall hydrolases responsible for this breakage act outside of the cell in a dense meshwork, physically removed from most potential mechanisms of regulation. How is their activity controlled?
First, we characterized the activity of the hydrolase enzymes. Using fluorescent D-amino acids (FDAAs) to measure wall turnover, we found that there are two classes of hydrolase enzymes with separable functions: one group removes of old material from the wall, and the other breaks bonds in the wall to allow for cell growth. Strikingly, were able to knock out all of the hydrolase genes except those responsible for growth (35 in total) with no change in viability or wall thickness.
Next, we investigated the regulation of these activities, and found that mechanical stress plays a key role. We used osmotic shocks to alter stress on the cell wall, and measured autolysis rates as a proxy for the activity of the hydrolases. These experiments indicate that the enzymes responsible for cell wall removal preferentially cleave material that is not under stress. FDAA and electron microscopy experiments indicate that the growth enzymes are also spatially regulated. We thus propose a stress-based model for regulation of wall thickness and growth.

#56
Hao Wu
GSAS
Chemistry and Chemical Biology

Olfactory Evidence Accumulation in Mice

We ask if mice can differentiate two trains of intermittent odor pulses with the same transient but different statistics. We investigated the neural activities of these mice during their learning and trained behavior in the olfactory bulb. Trained mice can readily differentiate discrete pulses arriving with Poisson statistics with number ratio 3:1 in a span up to 10s with greater than 80% performance.

#57
Min Wu
University Of Basel
Department of Environmental Sciences, Zoology and Evolution

The genetic mechanism of selfishness and altruism in parent-offspring coadaptation

The social bond between parents and offspring is characterized by coadaption and a balance between altruistic and selfish tendencies. Yet the underlying genetic mechanism remains unknown. Using transcriptomic screens in the sub-social European earwig, Forficula auricularia, we found the expression of over 1600 genes associated with parental care in mothers or offspring. Furthermore, we identified two genes, Th and PebIII that were synergistically up-regulated in mothers' heads and offspring during parent-offspring interaction. In vivo RNAi experiments confirmed direct and indirect genetic effects of Th and PebIII on behavior and fitness, including maternal food provisioning and reproduction, offspring development and survival. The direction of the effects in mothers and offspring consistently indicated a reciprocally altruistic function for Th and reciprocally selfish function for PebIII. Metabolic pathway analysis suggested roles for Th-restricted endogenous dopaminergic reward, PebIII-mediated chemical perception and insulin-signaling, juvenile hormone and vitellogenin interplay in parent-offspring coadaptation and social evolution.

#58
Xingbo Yang
FAS
Molecular and Cellular Biology

Biophysical studies of metabolic control in mouse oocytes and embryos

While a great deal is known about pathways and enzymology of carbohydrate metabolism, it is still poorly understood how metabolic fluxes are modulated during development and in response to environmental factors, or degraded in disease. Mounting evidence suggests that defects in metabolism may cause chromosome segregation errors in eggs and embryos, leading to age related infertility in women, but the possible underlying mechanisms remain unclear. This is partly due to the lack of approaches to measure metabolic fluxes in situ with subcellular resolution and the lack of a quantitative theory of metabolic control. We are developing and testing a coarse-grained biophysical model of enzyme engagement of electron carriers, with the goal of extracting metabolic fluxes from fluorescence lifetime imaging microscopy measurements. We are attempting to construct a theory of the control of mitochondrial respiration and cytoplasmic fermentation in mouse oocytes and embryos by using this approach in conjunction with metabolic manipulations. Our preliminary results argue that the fluxes through these pathways can be redirected in response to perturbations, but this is accompanied by large changes in cell biological features, including disassembly of the spindle, which we speculate might underlie the connection between metabolic defects and chromosome segregation errors. Our study provides a framework to probe interactions between different metabolic pathways in a quantitative manner that can help reveal new metabolic control mechanisms and their effects on cell and developmental biological processes.

#59
Liyuan Zhang
SEAS

Cell in Gel

Hydrogel is a water-swollen crosslinked polymer network that can provide spatial and temporal control over the release of various therapeutic agents, including small-molecule drugs, proteins and cells. Encapsulating stem cells in a biocompatible hydrogel provides tremendous potential for stem cell therapy. These encapsulated stem cells retaining their ability of human tissue regeneration revolutionize drug development, organ modeling and precision medicine. However, the stem cell encapsulation-based therapy depends on materials science. Stem cells in these hydrogel matrices encounter a different microenvironment compared with that of the true physiological conditions, and often result in reduced performance or loss of key functionality. Other important challenges remain in its clinical application, such as low cell viability after encapsulation and translation, and poor compatibility of the hydrogel, which hampers a broader application. Moreover, heterogeneity of the stem cells from patient to patient makes this an even more difficult problem. In this talk, I will focus on the technique we have developed for encapsulation and assembly stem cells by droplet microfluidic technology and how we control the stem cell behavior using various biomaterials with different geometry.

#60
Pengjuan Zu
MIT
Evolutionary biology

Chemical communication explains plant-herbivore interaction networks: an information theory approach

Through millions of years of arms races between plants and insect herbivores, plants have developed diverse chemical weapons against herbivores, which insects have evolved to cope with. Plant volatile organic compounds (VOCs) are one important chemical group in plant-herbivore communication. While functional roles of VOCs in plant-herbivore interactions have been acknowledged, few studies have investigated the informative roles of VOCs in shaping the interaction network. In this study, we employ information theory to build a novel conceptual framework of chemical communication to explain plant-herbivore interaction networks. Specifically, we model that both plants and animals aim to eavesdrop information to identify their opponents (i.e. increase mutual information of VOCs and the opponents), without leaking information of themselves (i.e. decrease corresponding mutual information to confuse the opponents). While the language is shared by both parties, the resulting trade-off becomes the underlying mechanism in driving a stable pattern of plant-herbivore chemical communication. We apply this framework to a well-studied plant-herbivore network in a Mexican tropical dry forest to test whether the observed plant-herbivore interactions will emerge under this theory and how the chemical information evolve with different network structures. This novel work opens a distinct and promising angle of understanding plant-insect communication

Export

iCal

QuantBio/NSF Simons Poster Session

Date:

Location:

Poster Summary

#

Author

Institution

Poster Title

Poster Abstracts

QBio mailing list

Thursday seminar mailing list