Poster 1: Role of mistranslation in bacteria-bacteriophage interactions

Aabeer Kumar Basu

The importance of variability within host populations in determining disease incidence and pathogen prevalence is well established. Increased host variability limits pathogen spread and prevents host extinction. Although most empirical and theoretical studies exploring this phenomenon focus on genetic variability within host populations, variability in susceptibility to pathogens can also come about due to non-genetic modifications of the host phenotype. For example, in invertebrates prior experience with a non-lethal infection is known to help the host survive a lethal infection with the same pathogen during a second exposure. The aim of our study is to explore the importance of phenotypic variability, in absence of an underlying genetic variability, in the host population in context of susceptibility to pathogens using a tractable bacteria-bacteriophage model system. Mistranslation, including incorporation of incorrect amino acids into polypeptide sequences, is a prevalent phenomenon in bacteria. Mistranslation increases phenotypic variability within clonal bacterial populations, helping the bacterium explore an extended range of phenotypic space without having to acquire the necessary mutations. Additionally, mistranslation leads to upregulation of baseline stress response levels in bacteria which helps it survive a variety of stressors at first exposure before adaptive evolution kicks in. In our experiments we tested if modifying the level of mistranslation in Escherichia coli changes its susceptibility to infection by bacteriophage P1. Our preliminary results suggest that modifying mistranslation rate in either direction, i.e., both increasing and decreasing the rate of mistranslation, can make E. coli resistant to phage P1, although the observed resistance is dependent on the mutation introduced to help change mistranslation rate. Increased resistance is apparent in both solid agar medium and in liquid cultures.

Poster 2: Plants' Phenotypic Plasticity in Response to Fly Ash

Anjali Verma

Plants are experiencing heterogeneity due to various environmental stressors like climate change, invasiveness, pollution, stress condition etc. Phenotypic plasticity is considered as one of the major means by which plants can cope up with environmental factor variability. One of the main causes of worry on a worldwide scale is environmental pollution brought on by the production of smoke, gases, and solid wastes like fly ash from thermal power plant(CPCB). Fly ash is a major problematic pollution. it can directly affect the health of plant species with greater adaptive plasticity. Fly ash amended with different organic and inorganic substances and their affect on plant species such as Glycine max L. Zea mays L. are different at morphological and physiological level in different doses. 20% of fly ash significantly increases plant biomass, flowering, root and shoot length but decreases all growth parameters in between 25% to 30% fly ash as soil amendment. These stochasticity of plant species has been proposed as a target for the development of more productive crops in variable amendment and edaphic stress conditions. This heterogeneity may facilitate survival under extreme environmental conditions. Species with a greater adaptive plasticity may be more likely to survive in varied environmental conditions. In the near future it will be important to increase long term studies on natural population in order to understand plant responses to different environmental factor.

Poster 3:

Anjoom Nikhat

Expression levels of different clock genes relative to the external light-dark cycle dictate the phase of the circadian clock, which differs among individual, as well as between tissues of the same individual. Application of chronomedicine requires estimation of this phase from minimum number of samples. However the low copy number associated noise in the expression of the clock genes makes phase prediction challenging. We perform multiplexed SABER-FISH to detect the absolute clock genes’ mRNA numbers in Dexamethasone synchronized monolayer of NIH3T3 over ~48hrs. We extend the previously developed oscillation detection package OdeGP to learn the oscillatory pattern of multiple clock genes using the collected time course data and then using Maximum likelihood estimate approach detect the circadian phase of a test sample. Our algorithm predicts the phase of the circadian phase with an error ~2hrs. Further in we see a decrease in error when we average the test data counts over increasing cell numbers, suggesting that circadian phase is an emergent property of group of cells rather than that of single cells. Using an information-theoretic approach we also try to quantify time information in bits associated with stochastic clock genes’ expression levels, to understand whether time is decoded post-transcriptionally.

Poster 4:

Anton Swaminathan Iyer

Isogenic cancer cells are known to develop non-genetic heterogeneities due to epigenetic modifications, protein/gene expression variations and post-translational modifications. These heterogeneities result in fractional death of such cells when exposed to a drug such as cisplatin. Identification of these heterogeneities is still a challenging problem in the field. In the present work we study the effect of this phenomenon on population dynamics using traditionally prevalent exponential growth models. Age structured population growth models are used to predict the growth rates of cells from single cell times to division and death, IMT and AT. Using single cell time lapse data for colorectal and osteosarcoma cell lines, we show these models accurately predict growth rates only in the drug naive cell population. For drug exposed cells, traditionally prevalent models show a large error in their estimates. Taking reference from previous work on correlations in fates of lineage related cells undergoing treatment, we construct minimalistic models (with increasing set of parameters) that utilize bivariate statistics to arrive at the simplest model that matches growth rate estimates and correlations in fates of lineage related cells. We further show that our model matches with complexity reduction in distinctly surviving lineages and in the persister cell distribution as seen in the experiment.

Poster 5: Combination of cellular rearrangement and Notch-Delta signaling drives patterning of utricle 

Anubhav Prakash

Every organ consists of different cell types, developed from precursor and patterned into a multi-cellular array. This emerges from cell growth, differentiation encoded by genetic programmes and the interplay of cells physical and biochemical properties. How cells integrate this information in a developing tissue is largely unexplored. We are investigating this question using the mouse utricle, where the organ is an epithelium composed of two cell types hair cells (HC) and supporting cells (SC). 

In development, from embryonic day (E) 13.5 to birth P0, the utricle increases in size and progenitor produces mosaics of HCs and SCs. The differentiation of HC and SC is known to be driven by notch-delta signalling, here delta-high cells differentiate into HCs and accentuate notch in neighboring cells, preventing their differentiation into HCs. Although, proliferation is over by birth, the number of HCs increases from 1500 to 3500 in the next 4 days, raising the question how HC number doubles with existing cells? 

We rationalise, the fate of SC to be plastic and competent to differentiate into HC in these 4 days. Indeed, experimentally we observed an increase in the number density of HCs and the subsequent decrease in number of SC. We found this differentiation of non-HCs into HC is driven by cellular rearrangements such that their HC contacts and hence notch-mediated attenuation is lost. Subsequently, the commitment to the SC fate in non-HCs increased. 
Our data suggest an unknown mechanism that integrates cues from Notch-Delta signalling and cell-contacts to develop mosaics of HCs and SCs in the utricle. 

Authors: Anubhav Prakash*, Raj Ladher, Sandeep Krishna (* Presenter)

Poster 6:

Arun G S

Hyperpolarisation NMR to probe conformational heterogeneity of biomolecules, and more. Inherently dynamic biomolecules exist in equilibrium with several of their conformations, and those have been studied at atomistic resolution through solution-NMR (Nuclear Magnetic Resonance) spectroscopy at their native physical state.

Nevertheless the low sensitivity of NMR meant that much of the functionally relevant lowly populated states remained inaccessible. Additionally, having to keep samples at higher concentration may lead to the creation of non-native conformational landscape. With much interest in recent years, researchers have been trying to enhance the nuclear spin-polarisation levels by several strategies, thereby increasing the NMR sensitivity several fold.

Active research is ongoing in the field, in developing techniques that enable efficient and stable hyperpolarisation of nuclei, in liquid as well as solid state NMR. Combined with in-cell NMR, hyperpolarisation techniques offer a non-invasive probe to follow cellular events at various timescales, and without having to alter the physiological concentrations of any biomolecule.

It would also mean a shift in the way we probe structure-function; from ‘extrapolating in-vitro data to native cellular context’ to ‘collecting data directly from native environment’. At SPLS23, I shall try to highlight various advancements in the field, and explore what hyperpolarisation NMR has to offer to biological sciences in probing “plasticity of structure” and more!

Poster 7:

Avijit Das

Poster 8: Neural Stem Cell-Specific Transcription Factors Specify Lineage Identity by Modulating Chromatin Accessibility

Ayanthi Bhattacharya

Poster 9: An evolutionary paradigm favoring crosstalk between bacterial two-component signalling systems

Bharadwaj Vemparala

The prevalent paradigm governing bacterial two-component signaling systems (TCSs) is specificity, wherein the histidine kinase (HK) of a TCS exclusively activates its cognate response regulator (RR). Crosstalk, where HKs activate noncognate RRs, is considered evolutionarily disadvantageous because it can compromise adaptive responses by leaking signals. Yet, crosstalk is observed in several bacteria.

To resolve this paradox, we proposed an alternative paradigm where crosstalk can be advantageous. We envisioned ‘programmed’ environments, wherein signals appear in predefined sequences. In such environments, crosstalk that primes bacteria to upcoming signals may improve adaptive responses and confer evolutionary benefits. To test this hypothesis,

We employed mathematical modeling of TCS signaling networks and stochastic evolutionary dynamics simulations. We considered the comprehensive set of bacterial phenotypes, comprising thousands of distinct crosstalk patterns, competing in varied signaling environments. Our simulations predicted that in programmed environments phenotypes with crosstalk facilitating priming would outcompete phenotypes without crosstalk. In environments where signals appear randomly, bacteria without crosstalk would dominate,

Explaining the specificity widely seen. Additionally, a testable prediction was that the phenotypes selected in programmed environments would display ‘one-way’ crosstalk, ensuring priming to ‘future’ signals. Interestingly, the crosstalk networks we deduced from available data on TCSs of Mycobacterium tuberculosis all displayed one-way crosstalk, offering strong support to our predictions. Our study thus identifies potential evolutionary underpinnings of crosstalk in bacterial TCSs, suggests a reconciliation of specificity and crosstalk, makes testable predictions of the nature of crosstalk patterns selected, and has implications for understanding bacterial adaptation and the response to interventions.

Poster 10: Shape-Shifters of the Skin: Dermal Fibroblast Plasticity and Functional Heterogeneity in Wound Repair and Skin Fibrosis

Gaurav Kansagara

Fibroblast activation is a key cellular process required for the proper healing of wounds, the pathogenesis of fibrotic diseases, and the tumor stroma. "Activation", however, is an umbrella term that includes fibroblast – proliferation, migration, contraction, inflammatory cytokine production, and excessive collagen secretion. This raises an interesting question of whether the same fibroblast performs all these functions or whether there are subpopulations of fibroblasts performing specific functions. To understand this, we use a Snail transgenic (Snail Tg) mouse model of dermal fibrosis (Rana et al., JID, 2022). Using this model, we have uncovered fibroblasts' cellular and molecular heterogeneity in wound healing that also fuels fibrosis when dysregulated.

Specifically, we have identified a protein, Mindin, secreted during wound healing and skin fibrosis, which has multiple effects on different fibroblast subpopulations. Mindin increases inflammatory cytokine production and migration of the reticular subpopulation of fibroblasts, a phenotype similar to inflammatory cancer-associated fibroblasts(iCAFs). On the other hand, the papillary fibroblasts respond to Mindin with elevated contractile activity and collagen secretion and adopt features of myofibroblastic cancer-associated fibroblasts(myCAFs). Furthermore, we have identified the specific members of Src family kinases and Rho GTPases, which mediate the unique downstream effects of Mindin on the different fibroblast subpopulations. Currently, our efforts are focused on identifying the receptor mediating the effects of Mindin on fibroblast subpopulations.

Our work presents better mechanistic insights into fibroblast heterogeneity and its role in wound healing and skin fibrosis. Fibrosis affects nearly every tissue in the body and contributes to 30% of deaths worldwide. Despite this enormous public health burden, there are currently no effective drugs available to combat fibrosis, and the discovery of these new pathways offers novel routes for therapeutic intervention.

Poster 11: Modeling cell fate determination by modulating the architecture of gene regulatory network and noise-assisted transitions between attractor states

J Hareesh

Despite having similar genetic compositions in newly differentiated cells, the phenotype is determined by the relative position of the cells in an organism and their environment. The ability of cells to adapt their fate in response to changes in the environment must be guided by various mechanisms, both direct and indirect in the gene regulatory network (GRN). To explore such mechanisms, we construct synthetic GRNs and tabulate the attractor states generated by the combinatorics of various gene expression profiles. For simplicity, the effect of environment, any non-monotonicity in the response, indirect regulation, cooperative and/or higher order effects are effectively accounted for, by using suitable motifs and additional proxy nodes. Using a variant of conventional spin models used in statistical physics that takes into account the directed nature of the network, we represent gene expression as a binary variable. The transition probabilities between attractor states are regulated by inhibitory feedback loops operating at a slower timescale and by modulating the noise (quantified by a “temperature” parameter). The connectivity of the transition matrix determines the resolution at which one is scanning the phenotypic space. By modulating the attractors that are accessible to a cell having a specific gene expression profile, the cell can perform an efficient search of accessible states in its neighborhood and converge to the state that is in accordance with its positional information. Suitable wiring of the inhibitory feedback loops allows us to encode a specific developmental trajectory and include the effect of inter-cellular interactions during development. Further, we study the robustness of these trajectories to ambient stochastic fluctuations. Understanding information processing in directed networks is non-trivial and the results have implications in structural analysis of GRNs.

Poster 12:

Jyotsana Jewel Parmar

For the transcription of several genes, the distal regulatory element known as enhancer must approach their respective gene promoter for the efficient and robust transcription output. Several transcription factors, co-regulators and mediators bind to the enhancer elements which then loop to their promoter sites in order to initiate transcription. In estrogen-positive breast cancer cells, following estrogen signaling, the transcription factor ER-alpha binds to its specific enhancer elements, thereby modulating the transcription of multiple genes that are otherwise in a distinct transcriptional state. Chromatin capture studies show that this process rearranges the chromatin landscape forming several new loops and transcription domains. Moreover, when visualized under the microscope, tagged ER-alpha shows several puncta/foci inside the nucleus whose number and size grow with time. However, the nucleation and growth dynamics as well as the spatial dynamics of these puncta remain unexplored. In this work, we perform a quantitative analysis of live cell imaging data of ER-alpha upon signaling. We show that the nucleation of ER-alpha is heterogeneous in space and the new nucleation events happen in the close proximity of the older clusters. Dynamical analysis shows a wide range of diffusive dynamics of these puncta, some moving fast and some in the sub-diffusive range.

Poster 13: Subtle alteration in transcriptional memory governs the lineage-level cell cycle duration heterogeneities of mammalian cells

Kajal Charan

Mammalian cells exhibit a significant level of intercellular heterogeneity in cell cycle period and phase durations even under culture conditions. However, distinct correlation patterns in cell cycle durations emerge at the lineage level among mother-daughter, daughter-daughter, and cousin-cousin cell pairs.The factors governing these correlation patterns, as well as the mechanisms of inherited cellular memory across generations, are poorly understood. Herein, we examine the simplest generic cell cycle model under the influence of correlated transcriptional fluctuations that illustrate how transcriptional memory inherited across generations can get altered during the cell cycle progression to produce a variety of lineage-level cell cycle duration correlation patterns. Our Model demonstrates the parametric space comprised of the intensity of the transcriptional fluctuations and their correlation timescale that reconciles the experimentally observed correlation pattern among various lineage cellular pairs for different cell types. Importantly, our model predicts that the extent of correlation among mother-daughter and cousin-cousin lineage pairs intricately outweighs each other depending on the cell cycle duration without much changes in daughter-daughter correlation. These insights will be crucial in tackling cellular heterogeneities in a cell-type-dependent manner in the future.

Poster 14: A trait-based framework to integrate resilience frameworks

Karthik Murth

In the past few decades, there has been an unprecedented interest in resilience studies across natural science and engineering disciplines. This has resulted in numerous frameworks that are highly context specific, with each framework based on a set of rules and assumptions, thus compromising the generality of its application.Our study focuses on the inherent traits of a system that aid in managing disturbances, thus deviating from the traditional mathematical frameworks in resilience studies. We identify two important traits of system, which either provide capabilities of withstanding disturbances or the ability to recover from these disturbances. A spring-damper mechanical system (SDMS) provides a suitable analogy, as the spring and damper represent recovery and withstanding capabilities respectively to an external force. We test the applicability of SDMS to represent ecological/biological systems by inducing bi-stability into the SDMS equation and develop an integrated resilience index (IRI) that can distinguish systems based on their resilient properties. We examine the applicability of IRI in two scenarios: 1) systems with differential resilience properties experiencing similar external disturbances 2) systems with differential resilience properties experiencing varying intensity of external disturbances. Numerical simulations were conducted to examine SDMSs and mathematical models representing ecological systems for these scenarios. IRI was successful in distinguishing systems based on their resilient properties. Importantly, the formulation of IRI is based on the long-term response of the system to a continuously varying external disturbance and thus, overcomes limitations of the traditional resilience frameworks, particularly in the context of engineering and ecological resilience frameworks applied in ecology. Although the modeling examples are motivated by ecological or biological systems, this novel approach to measure resilience is simple and versatile, thus opening opportunities to be tested on empirical datasets of diverse disciplines.

Poster 15: Heterogeneity: Shaping The Survival In Biofilms

Kavi Bharathi R

Biofilms are highly heterogeneous social entities that transform independent cells into specialized cell populations. Biofilms are essential for establishing the infection and bacterial pathogenicity in the host.These structurally heterogeneous groups undergo spatiotemporal changes correlating with the matrix gene expression. Spatial organization within the biofilm community is assigned in such a way to benefit the residents to reduce the overall cost and benefit from the common goods while being protected from the exogenous environmental insults. We studied two bacterial groups of the Enterobacteriaceae exhibiting heterogeneity to enhance their overall robustness. In the first group, Pellicle biofilm of Uropathogenic Escherichia coli plated on the Congo red agar, produced red (RCV), white (WCV), and hetero (HCV) colony variants. One of the key elements determining the population's life and existence is the sugar source in the media, The swarming ability of these variants in co-culture to sequester glucose and their social interactions post-uptake was analyzed using fluorescence microscopy followed by image analysis using ImageJ and DAIME (Digital Image Analysis in microbial ecology). Further RNA sequencing was performed to comprehend the differential gene expression of the genes regulating these heterogeneous populations. The second group Salmonella produced colony, submerged, and pellicle biofilms in varied interfaces of solid, liquid, and air. We observed heterogeneity within each model in a single biofilm, their resilience to stress exposure, and how one matrix producer is counterbalancing the other to survive stress conditions to establish themselves as a single community was also studied using the co-culture. Interestingly these observations show how these heterogenous populations efficiently compensate for each other's abilities adeptly in establishing themselves in the host and infectious conditions.

Poster 16:

Kritika Kumari

The intracellular Vesicle Traffic system is a defining feature of Eukaryotic cell organization. It consists of vesicle budding from a source endomembrane compartment and fusion with specific target(s) compartment mediated by molecular interactions of target and vesicle molecules. Therefore, the transport of cargo across cellular locations is regulated by molecular interactions, which in turn depends on how these molecules in play are regulated. A network of vesicle trafficking in a homeostatic state has been established using the Rothman-Schekman-Sudhof[1] model, and the constraints leading to certain properties of the emerging network have been encoded by Graph-theoretic modelling[2-3] and model checking through Boolean Satisfiability(SAT) [2]. These Graph encoding of networks represents compartments as nodes and vesicles transported in-between as directed edges. Some of the basic constraints upon which these Graphs are generated include the steady state of the compartments requiring cyclical transport of cargo and the molecular interaction specificity-modules directing the edges to their target. These constraints onlyallow Graphs with certain properties like Graph Connectivity to be relevant in the biological framework.

My work focuses on investigating how viral hijack of endomembrane traffic can be studied using this model, and discovering the new constraints arising from their introduction into the existing allowed vesicle traffic transport graphs. Viruses are known to use endocytic vesicular path into the Vesicle Traffic network evading cytosolic components while exploiting endocytic maturation[5] and reorganization of compartments to form Replication Organelles(ROs)[4]. Endomembrane traffic system also enables the main stages of the viral life cycle, i.e., entry, replication, assembly, and egress of progenies from the host cell and immune evasion [5]. Thus, our focus is on viruses that subvert the host vesicle cellular trafficking. We aim to realize how viruses can perturb this network in terms of Graph connectedness, the reachability of different solution graphs via molecular or functional trasitions, i.e., changes in the steady-state dynamics of a cell.

References:

1. Rothman JE. (2002). ”The machinery and principles of vesicle transport in the cell.” Nature Medicine, 8(10),p.1059–1063. https://doi.org/10.1038/nm770

2. Shukla, A. et al. (2017) “Discovering vesicle traffic network constraints by model checking,” PLOS ONE, 12(7), p. e0180692. https://doi.org/10.1371/journal.pone.0180692

3. Mani, S., Krishnan, K. Thattai, M. (2022) “Graph-theoretic constraints on vesicle traffic networks”. Journal of Biosciences, 47(11). https://doi.org/10.1007/s12038-021-
00252-5

4. Hernandez-Gonzalez, M., Larocque, G. and Way, M. (2021) “Viral use and subversion of membrane organization and trafficking,” Journal of Cell Science, 134(5).
https://doi.org/10.1242/jcs.252676

5. Cossart, P., Helenius, A. (2014). “Endocytosis of Viruses and Bacteria”. Cold Spring Harbor Perspectives in Biology, 6(8). https://doi.org/10.1101/cshperspect.a016972

Poster 17:

Nandita Chaturvedi

The question of adaptation to environmental change at long time scales remains an important area of theoretical investigation within evolution and ecology, but has been treated mainly in the context of single environments. However, organisms almost always deal with multiple environments and trade offs arising from them, in addition to the possibility of long term change. Here we combine the idea of repeated variation or heterogeneity, like seasonal shifts, with long term and directional dynamics. To address this complex situation, we extend the framework of fitness sets and study how the optimal phenotype in this situation can itself change with long term shifts. We consider selection from two distinct environments. We find that the behavior of a population under such a system is qualitatively different and more complex than that of a population responding to long term change in a single environment. The chance of survival or extinction depends crucially on the relative frequency of the two environments, the strength and asymmetry of their selection pressure in addition to population size, phenotypic diversity, fecundity and the rate of change of the environment. We study characteristics of the population under selection such as its phenotypic lag behind the optimal phenotype and the mean time to extinction.

Poster 18:

Priyotosh Sil

Boolean network (BN) models of gene regulatory networks (GRNs) have gained widespread traction as they can easily recapitulate cellular phenotypes via their attractor states. Their overall dynamics are embodied in a state transition graph (STG). Indeed, two Boolean networks (BN) with the same network structure and attractors, could have drastically different STGs depending on the type of Boolean functions (BFs) employed. A key objective of this work is to systematically delineate the effects of different classes of BFs on the structural features of the STG of reconstructed Boolean GRNs while keeping network structure and biological attractors fixed, and explore the characteristics of BFs that drive those features. For that, we draw from ideas and concepts proposed in cellular automata for both the structural features and their associated proxies. We use the network of 10 reconstructed Boolean GRNs to generate ensembles that differ in the type of logic used while keeping their structure fixed and recovering their biological attractors, and compute quantities associated with the structural features of the STG: ‘bushiness’ and ‘convergence’, that are based on the proportion of garden-of-Eden (GoE ) states and transient times to reach attractor states when originating at them. We find that using biologically meaningful BFs lead to higher ‘bushiness’ and ‘convergence’ than random ones. Computing these ‘global’ measures gets expensive with larger network sizes, stressing the need for more tractable proxies. We thus generalize Wuensche’s Z-parameter to BFs in BNs and provide 4 variants, which along with the network sensitivity, comprise our descriptors of local dynamics. Certain network Z-parameters and the network sensitivity were found to be very good proxies for the bushiness. Finally, we provide an excellent proxy for the ‘convergence’ based on computing transient lengths originating at random states rather than the GoE states.

Poster 19: Unravelling the Heterogeneity of HUMAN CD8 T Cell responses: Implications for Vaccine studies

Rama Akondy

Cytotoxic T cells (referred to here on as CD8 T cells) are a part of the adaptive immune response that play a pivotal role in controlling intracellular pathogens and tumors. However, eliciting a CD8 T cell response through infection or vaccination does not always guarantee complete protection. Understanding the characteristics of functional CD8 T cells and their identification remains a challenge. This is especially true in human studies, where we depend on clues from blood-derived cells to gain insights into CD8 T cell behavior. In the past, human CD8 T cells have been categorized into four phenotypic subsets based on the expression of the cell surface proteins CD45RA and CCR7.  While this classification was a step in the right direction, various high dimensional, single cell analyses in the past decade reveal that the heterogeneity observed in human memory CD8 T cells extends well beyond these four subsets. Here, we address the nature of heterogeneity within human CD8 T cells, the plasticity of the phenotypes present and the need for cautious interpretation based on what memory CD8 T cells ‘look’ like. There are important implications of these new insights for vaccine evaluation.

Poster 20: Effect of Heat Stress on Inter Cellular Variability

Rituparna Goswami

Plants with the identical genotype growing in similar macroenvironment exhibit gene expression variability, particularly enriched for biotic and abiotic stress response pathway genes (Cortijo et al., 2019). Due to the sessile nature of plants, temperature fluctuations become an unavoidable environmental feature which plants need to anticipate and respond to. Plants have evolved elaborate Heat Shock Response (HSR) molecules and pathways. HSP70 (a chaperone protein) is an important downstream molecule in HSR pathways and shows high levels of inter-plant variability in gene expression immediately after dawn. To further zoom into this expression variability at the cellular level HSP70::mTurquoise-N7 transcriptional reporter lines have been imaged under mock and heat stress conditions. In unstressed plants, we observed heterogeneous expression of HSP70 in individual cells with only a small number of cells showing expression. There was also inter-plant variability, with the amount of ON cells varying from plant to plant.
This pattern varies between different tissues like root tip and hypocotyl, where mature tissue like the hypocotyl has more spread-out profile of gene expression in individual cells in the unstressed state. Depending on duration of the 35°C HS, we observe dose dependent change in intercellular expression pattern of HSP70::mTurq-N7 expression. Even 24 hours after the stress, some cells remained OFF, although the number of cells with high expression was larger compared to unstressed plants. Both hypocotyls and roots show increase in the spread of cellular level gene expression. This preliminary data portrays a scenario where in response to HS, rather than a gradual increase in all the cells, individual cells show noisy activation of transcription and noisy change in the level of expression of HSP70 expression. Also, it suggests inter-individual variability is caused by individual plants having different amounts of variably expressing cells. Thus, we are beginning to reveal links between inter-cellular variability and inter-individual variability in plants.

Poster 21: Molecules to Graphs : Exploring the expansion of the eukaryotic vesicle traffic networks

Sahana K Shridhar

The vesicle traffic system is a defining feature of eukaryotic cells. The vesicle traffic system has been pivotal in not only the diversification of eukaryotes but is also central to cellular functions. In my work, I model the eukaryotic vesicle trafficking system using graph theory to understand the role of molecular interactions (genotype space) in shaping the structure and functionality of vesicle transport networks (phenotype space), and how molecular changes in an evolutionary time scale pave way for the emergence of new trafficking pathways, cellular functions and organelles in an existing , stable network.

We have modeled the molecular rules and constraints to first identify biologically relevant, stable transport graphs. We then explore the space of these allowed transport graphs to understand how one transport graph, representing a particular configuration of vesicle traffic, can transform into another graph due to loss/gain or duplication of the underlying molecular interactions . These transitions between networks are key to exploring how the architecture of the network adapts to new cellular functions or the optimization of existing processes (plasticity of networks) , and to further understand the possible evolutionary trajectories of the extant vesicle traffic system.

We aim to use our model to make predictions regarding essential features of the vesicle traffic networks, recycling pathways for trans-membrane molecules such as SNARE, and the emergence of non-endosymbiotic organelles in the proto-eukaryotic organism etc.

Poster 22: Ribosome availability connects transcriptional bursts with protein expression noise

Sampriti Pal

The existence of noise in gene expression is responsible for the emergence of phenotypic heterogeneity among individuals within a population. This noise plays a crucial role in various biological processes, from antibiotic persistence to anti-cancer therapy resistance. The preliminary research on noise uncovered many molecular characteristics linked to transcription that play a crucial role in regulating noise levels. Nevertheless, the noise produced during the transcription process is also influenced by translation, affecting the quantity of noise in protein expression. A limited study has been conducted to date on the impact of the translation process on noise. Those studies have found positive correlations between translational efficiency and protein noise in multiple species. Although certain pathways have been hypothesized, the molecular basis of such positive correlation remains unclear. Using mathematical translation modeling at the level of a single mRNA and random simulations, we find that changes in ribosome availability as a major driver of translational modulation of transcriptional bursts. This model gives rise to two predictions which we experimentally confirm through a protein noise measurement assay. Here, we empirically integrated a promoter-fluorescent gene construct in the genome of yeast Saccharomyces cerevisiae and introduced coding region mutations in the fluorescent gene to quantify their impact on noise through single-cell flow cytometry data. Most uniquely, this study shows that genes with low mean mRNA expression and high translational efficiency can lead to high protein expression noise. Our result provides a general molecular mechanism behind the positive correlation observed between translational efficiency and protein noise.
 

Poster 23:

Shubham Shinde

The lower airway epithelium consists of mainly club cells CCs (secretory cells) and ciliated cells with rare variant club cells (variant-CCs) in specific regions in the tissue. The balance between the CCs and ciliated cell population is essential for proper functioning of the the lung. We study the dynamics of the proportions of CCs vs club cells upon two different perturbations to the tissue viz., pharmacological and genetic. CCs have the ability to self-renew and upon Notch inhibition they transdifferentiate to become ciliated cells. However, variant-CCs transition into an ambiguous state and upon restoration of Notch signalling these cells revert to become CCs. Thus, variant-CCs act as a source to replenish the CCs and ciliated cells. The transition of undifferentiated cell into either CC or ciliated is modeled after Corson 2019 as a decision along the unstable manifold of the unstable fixed point. Saddle node bifurcation controlled by Notch signalling takes the undifferentiated cell either to become a CC or a ciliated cell. With 1D simulation, the proportion of CCs as a function of tissue size is studied for different parameters such as signalling range, signalling strength, and choice of signalling function. The model fits well with the experimental data for the notch inhibition case for a signalling strength of slightly more than one cell length. One of the mechanism with which the signalling could be more than cell length is the use of cell protrusions to communicate.The geometry at the tissue level with current model is high dimensional, in the sense the homogeneous fixed point has many unstable directions. We are looking at possibility of low dimensional representation of the current dynamics.

Poster 24: Spatio-temporal phenotypic heterogeneity dictates removal of aberrant cells from epithelial tissue

Sindhu M

Genotypically similar cells of a tissue show phenotypic heterogeneity arising from differential gene expression, and asynchronized cell cycles. In epithelial cells, this spatio-temporal biochemical heterogeneity is complexed by coupling with cell-cell and cell-substrate forces. Such bio-physical heterogeneity in epithelia in space and time and its functional relevance remains uncharacterized. We study the link between the dynamics of mechano-sensing elements of the cells and cellular forces, both in space and time. We then show how a crucial epithelial function- recognition and removal of aberrant cells from the tissue, depends on maintaining the spatio-temporal bio-physical heterogeneity. Similar to previous reports, we found spatial hotspots of cooperative cellular forces dictated by jammed and unjammed populations. Interestingly, the extent of jamming and the force correlation length also govern the clustering of cells based on protein expression patterns, suggesting a crosstalk between the physical and biochemical spatial heterogeneity. By treating cells with contractility-modulating and noise-enhancing drugs, we modify biochemical heterogeneity, which in turn affects the physical heterogeneity. On the contrary, by modulating substrate stiffness, we modulate physical heterogeneity and in turn the biochemical heterogeneity. To understand if this heterogeneity might impact tissue defence against aberrant cells, we cultured mutant cells with healthy wildtype cells and found that the efficiency of mutant extrusion is dictated by spatial heterogeneity. We are now studying the evolution of spatial heterogeneity in actin levels and traction forces in the timescale of days to understand if changes in spatial heterogeneity over time might also affect mutant cell extrusion.

Authors: Sindhu M, Tanishq Tejaswi, Medhavi Vishwakarma

Poster 25: Unravelling the dynamics of cellular response to different environmental conditions through Quantitative Phase Imaging

Sreeprasad Ajithaprasad

Quantitative phase imaging (QPI) stands as a powerful non-invasive technique for monitoring crucial cellular parameters such as dry mass, optical volume, and cell area in vitro. Integrating this tool with an efficient cellular image analysis framework such as microfluidic single-cell imaging allows for the quantitative assessment of multiple cellular properties over multiple generations in a controlled environment. In our research, we employ this system to study optical thickness and optical volume heterogeneity within A549 and U2OS cancer cells. By subjecting these cells to specific environmental conditions and monitoring over successive generations, we can unveil differences in their responses. Particularly, our focus centers on quantifying variations in intracellular dry mass and the mobility of these heterogeneous cells over extended durations. Such a study promises insights into differences contributing to varying phenotypic traits exhibited under the influence of different controlled medium conditions. Furthermore, our investigation extends to explore the potential evolution of these cells over time, shedding light on how their properties may adapt and change in response to altered conditions such as exposures to antibiotic drugs. This comprehensive approach not only advances our understanding of cellular dynamics but also carries substantial implications for diverse fields, including cancer research and drug development.

Poster 26: Unravelling functional heterogeneity of cell populations in tumor microenvironments by multi-task evolution theory

Subhasis Datta

Tumor microenvironment contains heterogeneous cell population that perform different functions. These functions influence cancer progression and metastasis, and can modulate response to anti-cancer therapy. Earlier studies in patients have indeed shown variation in functional specialization across different patients that have clinical relevance, suggesting that the functional variation is often critical for deciding tumor progression. However, whether such functional variation exists within a tumor and whether emergence of certain functions can predict disease prognosis remain to be tested. In this work, we apply multi-task evolution theory to uncover functional heterogeneity within a tumor from single-cell RNA sequencing data. We ask whether specific molecular functions emerge with cancer progression and whether functional heterogeneity itself increases with progression. This analysis reveals different tasks being performed by a cell population, as well as identify the groups of cells associated with those tasks. In addition, this analysis enables us to study cellular transitions from one task to another. Preliminary analysis reveals that there is considerable functional heterogeneity within tumor cells and there is increase in heterogeneity with progression. Further analysis will help us understand the functional transitions occurring inside a tumor and the molecular basis of these transitions.

Authors: Subhasis Datta1, Riddhiman Dhar*

Poster 27:

Suvranil Ghosh

Gene expression memory refers to the timescale of a specific expression state that persists within an individual cell or a cell lineage. 'Slow' genes (fluctuation time scale longer than cell division ) might be significant due to their potential role in drug resistance in cancer and bacterial infection. Currently, proposed methods for identifying such genes require lineage information or multiple samples. Again, the current method for measuring gene expressions at the single cell level depends on single time point observation. These things make identifying memory genes difficult in vitro settings. Using a stochastic model of gene expression dynamics, we show that the existence of memory genes leads to a power law distribution in the eigenvalue spectrum of the covariance matrix for gene expression. We use this observation to formulate a potential algorithm for identifying memory genes within single-cell RNA-seq (scRNA seq) datasets. Current experimental techniques' limitations make scRNA seq datasets very sparse and noisy. It also requires different normalization techniques to take care of sequencing depth, library size, and other technical factors (which are different sources of unwanted covariance).

Poster 28: Topology and dynamics of gene regulatory network driving phenotypic heterogeneity within cancer cell population

Upasana Ray

Regulation of gene expression by transcription factors (TFs) creates complex networks known as gene regulatory networks (GRNs). These networks are dynamic in nature and can impact cellular traits in response to both internal cellular state and external environment. This phenotypic variability plays an important role in development of multicellular organisms, immune responses against pathogenic infections, tumor progression and anti-cancer therapy resistance. In multicellular systems, different cell types in a tissue can show varying levels of regulatory activity, which can lead to phenotypic heterogeneity within the tissue. Thus, by mapping out the GRNs and understanding the role of TFs’ activity in these networks, we can gain deeper insights into disease progression and develop more effective treatments. In our study, using single-cell RNA-sequencing data from lung cancer patients, we have constructed the active GRNs in diverse cell types within tumors and have estimated the dynamic activity of TFs that are part of these networks. In addition, we have analyzed the variation in topology and dynamics of the network across different stages of lung cancer and identified TFs that have crucial roles in disease progression. Furthermore, our analysis has also revealed intra-tumoral variations in GRNs among different cell types, which could contribute to phenotypic heterogeneity and can have important implications for cancer growth and therapy resistance. Our findings provide key insights into cancer progression and intra-tumor heterogeneity, which have significant clinical implications.

Keywords: phenotypic heterogeneity, gene regulatory network, topology, regulatory dynamics

Authors: Upasana Ray, Adarsh Singh, Debabrata Samanta, Riddhiman Dhar*

Poster 29:

Vinoth M

Cells switch ON or switch OFF genes by altering the state of histone modifications at specific regulatory locations along the chromatin polymer. These ON/OFF processes are carried out by a network of reactions involving enzymes. Histone modification marks spread to neighboring domains in the presence of enzymes. This spreading has been primarily studied as a kinetic process and in some cases, as a pseudo-spatial accounting for the contacts between the neighboring domains. In our work, we present a model considering diffusing enzyme-like particles that can react and spread the histone modifications. We show how the results of a reaction-diffusion model differ from that of a kinetic model. We show how a confined modification domain emerges without explicitly considering a nucleation site and how important it is to consider the enzyme limitation, which is implicit in our model.

Poster 30:

Yuuki Matsushita

Cell fate decisions at phenotypic level can be thought of (and modelled mathematically) as very few sets of simple decisions. The qualitative features of these decision trees could look similar across different cell fate decisions. In this sense, the process of cell differentiation is said to be universal. I want to explore if cellular reprogramming is also universal. In cellular reprogramming, almost all epigenetic memories of differentiated cells are erased by the over-expression of a few genes, resulting in regaining pluripotency, the potential for differentiation. Through numerical simulations, we have found a way to represent both cell differentiation and reprogramming in a universal form. I will also talk about similar approaches in some problems in evolution which I am going to study in near future.

Poster 7: Neural Stem Cell-Specific Transcription Factors Specify Lineage Identity by Modulating Chromatin Accessibility

Ayanthi Bhattacharya