Abstracts - Information Theory in Life Sciences, 22nd April 2024

10:00 Arrival and coffee in Mathematical Sciences Atrium

10:30 Jonathan Wattis (University of Nottingham)
"Information Theory for Identifying Phenotype-Genotype Relationships"

Abstract: We discuss how information theory can be used to identify which mutations of the genome correlate with variations in phenotype.  We discuss how significance levels can be determined, and how a mathematical models of the underlying algorithm relates to other stochastic models.  We also discuss how information theory enables correlations between different mutations to be identified.

11:15 Andela Markovic (UCL)
"Dynamics of positional information in the vertebrate neural tube"

Abstract: In developing embryos, cells acquire distinct identities depending on their position in a tissue. Secreted signaling molecules, known as morphogens, act as long-range cues to provide the spatial information that controls these cell fate decisions. In several tissues, both the level and the duration of morphogen signaling appear to be important for determining cell fates. This is the case in the forming vertebrate nervous system where antiparallel morphogen gradients pattern the dorsal-ventral axis by partitioning the tissue into sharply delineated domains of molecularly distinct neural progenitors. How information in the gradients is decoded to generate precisely positioned boundaries of gene expression remains an open question. Here, we adopt tools from information theory to quantify the positional information that neural cells receive and investigate how temporal changes in signaling influence patterning precision. The results reveal that the use of signaling dynamics, as well as signaling level, substantially increases the precision possible for the estimation of position from morphogen gradients. This analysis links the dynamics of opposing morphogen gradients with precise pattern formation and provides an explanation for why cells rely on time-varying signals to impart positional information.

11:35 Robert Insall (UCL and University of Glasgow)
"Chemotaxis - why we need information theory and what we can learn"

12:20 Lunch buffet and posters in Mathematical Sciences Atrium

13:10 Thomas Ouldridge (Imperial College London)
"Long-lived memories in biochemical systems: Uses and thermodynamic consequences"

Abstract: Paradigmatic biochemical systems, from kinase signalling networks to the polymer templating processes of the central dogma, copy and store information in (meta)stable molecular states. These molecular memories persist beyond the timescale of coupling to the original molecular data, which is both essential to their functionality and of fundamental thermodynamic significance. In this talk I will both explore the functionality of long-lived memories in biochemistry, and discuss the thermodynamic costs associated with these far-from-equilibrium, low entropy states. I will also consider the challenges associated with constructing synthetic molecular memories that function in this way.

13:55 Jeremy Guntoro (Imperial College London)
"Error Rate and Entropy of Correlated Heterogeneous Polymer Copies"

Abstract: Templated polymer copying, where the sequence of an information-carrying polymer is copied into another polymer, is the underlying principle behind all known methods of genetic information transfer, namely replication, transcription and translation. While there is considerable interest in identifying fundamental properties of polymer copying, earlier work was hampered by a lack of consideration for copy-template separation. It is now understood that copy-template separation is a necessary feature of templated polymer copying systems, with significant ramifications on their practical and theoretical complexity. Given the importance of copy-template separation, polymer copying models have been devised with separation at their core, leading to significant insights on the non-equilibrium nature of copying. However, existing theories are only valid for homogeneous copying, where individual monomer units are equivalent and polymers are composed of two kinds of monomers, or for copying systems where the rates of monomer addition are independent of previously added monomers. Building on existing work, we developed an exact method that allows for the evaluation of polymer distributions generated during the copying of finite heterogeneous templates (that is, when monomers do not have identical kinetic and thermodynamic properties) where rates of monomer addition may be dependent on the previously added monomer, thereby avoiding the need for lengthy simulations. We then used this method to investigate the effect of heterogeneity on polymer copying error rates for various parameter regimes. Finally, we used the obtained insights to identify a parameter space where heterogeneity in the monomer kinetics and binding energies can be used to significantly reduce copying errors relative to the content-weighted mean error of homogeneous copying using each of the two constituent monomers.

14:15 Benjamin Qureshi (Imperial College London)
"Equilibrium-like bound on information propagation in far-from-equilibrium molecular templating."

Abstract: Biochemical systems often use catalytic templates to selectively produce multiple distinct products from a shared set of ingredients. For example, proteins are copolymers created from the same set of amino acids; the formation of specific proteins is templated by mRNA molecules, which act catalytically. The set of pathways by which these products are created is often complex, involving motifs such as kinetic proofreading. Moreover, in many cases of interest, biological systems continuously turn over the products, recycling the basic ingredients for subsequent use. The net result is a distribution of products maintained by a non-equilibrium cycle; accurate copying of a small number of templates will lead to a sharply peaked distribution with low entropy.
The fundamental thermodynamics of these far-from equilibrium information-processing systems is poorly understood. Previous studies have generally focused on models of a single templated copolymerisation process, rather than considering the full non-equilibrium cycle. Here, we consider the fundamental thermodynamics of a class of systems in which a distribution of products is created and maintained by a set of catalysts that interconvert molecules between pools of inputs and pools of products. These interconversion processes can be arbitrarily complex, but the essence of templating means that equivalent pathways with the same overall free energy change exist for all possible products; products can only be differentially produced due to differences in the speed traversing these paths. Different pathways leading to the creation of each product are associated with different free-energy changes. We show that the largest difference between these free energies sets a lower bound on the Shannon entropy, of the product distribution, or an upper bound on the channel capacity if the molecular system is interpreted as an information channel. This free-energy difference also bounds the maximum probability of a single product in a system. Surprisingly, however, the distribution with minimum entropy is in general different from the one which maximises the probability of a single product.

14:35 Hardik Rajpal (Imperial College London)
"Emergence of Higher Order information structures in evolution"

Abstract: At what level does selective pressure effectively act? When considering the reproductive dynamics of interacting and mutating agents, it has long been debated whether selection is better understood by focusing on the individual or if hierarchical selection emerges as a consequence of joint adaptation. Despite longstanding efforts in theoretical ecology there is still no consensus on this fundamental issue, most likely due to the difficulty in obtaining adequate data spanning sufficient number of generations and the lack of adequate tools to quantify the effect of hierarchical selection.
Here we capitalise on recent advances in information-theoretic data analysis to advance this state of affairs by investigating the emergence of high-order structures — such as groups of species — in the collective dynamics of the Tangled Nature model of evolutionary ecology. Our results show that evolutionary dynamics can lead to clusters of species that act as a selective group, that acquire information-theoretic agency. However, this higher order organization breaks down during the transition region close to the error-threshold, where single species gain higher information-processing abilities.}
Overall, our findings provide quantitative evidence supporting the relevance of high-order structures in evolutionary ecology, which can emerge even from relatively simple processes of adaptation and selection.

14:55 Coffee

15:20 Barbara Bravi (Imperial College London)
"Statistical learning approaches to modelling the immune system"

16:05 Panagiota Kyratzi (Sorbonne, Paris) and Cyril Rauch (Univ. of Nottingham)
"Identifying the association between phenotypes and genotypes"

16:40 Joseph Egan (UCL)
"An information-theoretic model of receptor-ligand binding and its application to T cell activation."

Abstract: The reversible binding of T cell receptors (TCRs) to peptide-major histocompatibility complex (pMHC) ligands is the key event that initiates the adaptive immune response. As TCRs and pMHC ligands are often only present at low copy numbers their interactions are inherently stochastic, yet the role of stochastic fluctuations on T cell function is unclear. Taking a stochastic perspective, the entropy rate of the TCR-pMHC binding dynamics can be interpreted as the rate at which peptide-specific information is transmitted to the T cell from the encounter. This view provides an information-theoretic interpretation of T cell activation that explains a range of experimental observations.  Furthermore, a simple model that accounts for serial TCR-pMHC engagement, reversible TCR conformational change and TCR aggregation shows that such mechanisms can produce an intracellular signalling rate that approximates the entropy rate. Based on this analysis, it is proposed that the immune response is not initiated by the strength of binding between the T cell and the APC cell per se, but rather by the entropy rate. Thus, effective T cell therapeutics may be enhanced by optimizing the entropy rate of the TCR-pMHC interactions via appropriate modification of their binding and unbinding rates.

17:00 Sebastian Dohnany, Harriet Smith and Josh Bourne (The Francis Crick Institute)
"I like to move it: Informational measure of sensorimotor coupling underlying active perception in artificial and real cells"

Abstract: Cells process information from the environment in order to respond to stimuli and perform their required functions. We hypothesise that the sensory interface of a cell is dynamically adapted by movement-driven changes in cell shape. This idea, known as sensorimotor coupling, is the basis of active perception, a widely-studied process from the fields of neuroscience, robotics and psychology known to enhance decision-making and other cognitive tasks. We use information theory as a unifying quantitative framework to map these concepts, through the lens of basal cognition, to the single cell scale and study the role of sensorimotor coordination in cellular-level perception, in silico and in vitro. We have first designed and implemented a predictive, agent-based computational model of a single cell capable of complex shape changes. By calculating the flow of information, using the transfer entropy between cell membrane length and displacement of a nuclear marker of whole-cell activation, we show that we can measure the degree of sensorimotor coupling using cell-level variables. Moreover, by measuring the same cell-level variables in endothelial cells in vitro, we present preliminary evidence that this framework is applicable in biologically realistic scenarios. Our work highlights the importance and potential of using information theory to quantitatively investigate the role of dynamic morphological changes in cell sensing and to understand how cell motor dysfunctions contribute to tissue development and disease.

17:20 Close