arXiv subjects

Ilya Nemenman

Publications and source records attributed to Ilya Nemenman.

At least 73 records · Page 4Linked to original sources

arXiv2014

Efficient inference of parsimonious phenomenological models of cellular dynamics using S-systems and alternating regression

The nonlinearity of dynamics in systems biology makes it hard to infer them from experimental data. Simple linear models are computationally efficient, but cannot incorporate these important nonlinearities. An adaptive method based on the S-system formalism, which is a sensible representation of nonlinear mass-action kinetics typically found in cellular dynamics, maintains the efficiency of linear regression. We combine this approach with adaptive model selection to obtain efficient and parsimonious representations of cellular dynamics. The approach is tested by inferring the dynamics of yeast glycolysis from simulated data. With little computing time, it produces dynamical models with high predictive power and with structural complexity adapted to the difficulty of the inference problem.

q-bio.QM↗

arXiv2014

Automated adaptive inference of coarse-grained dynamical models in systems biology

Cellular regulatory dynamics is driven by large and intricate networks of interactions at the molecular scale, whose sheer size obfuscates understanding. In light of limited experimental data, many parameters of such dynamics are unknown, and thus models built on the detailed, mechanistic viewpoint overfit and are not predictive. At the other extreme, simple ad hoc models of complex processes often miss defining features of the underlying systems. Here we propose an approach that instead constructs phenomenological, coarse-grained models of network dynamics that automatically adapt their complexity to the amount of available data. Such adaptive models lead to accurate predictions even when microscopic details of the studied systems are unknown due to insufficient data. The approach is computationally tractable, even for a relatively large number of dynamical variables, allowing its software realization, named Sir Isaac, to make successful predictions even when important dynamic variables are unobserved. For example, it matches the known phase space structure for simulated planetary motion data, avoids overfitting in a complex biological signaling system, and produces accurate predictions for a yeast glycolysis model with only tens of data points and over half of the interacting species unobserved.

q-bio.QM↗

arXiv2014

Millisecond-scale motor encoding in a cortical vocal area

Studies of motor control have almost universally examined firing rates to investigate how the brain shapes behavior. In principle, however, neurons could encode information through the precise temporal patterning of their spike trains as well as (or instead of) through their firing rates. Although the importance of spike timing has been demonstrated in sensory systems, it is largely unknown whether timing differences in motor areas could affect behavior. We tested the hypothesis that significant information about trial-by-trial variations in behavior is represented by spike timing in the songbird vocal motor system. We found that premotor neurons convey information via spike timing far more often than via spike rate and that the amount of information conveyed at the millisecond timescale greatly exceeds the information available from spike counts. These results demonstrate that information can be represented by spike timing in motor circuits and suggest that timing variations evoke differences in behavior.

q-bio.NC↗

arXiv2013

Genotype to phenotype mapping and the fitness landscape of the E. coli lac promoter

Genotype-to-phenotype maps and the related fitness landscapes that include epistatic interactions are difficult to measure because of their high dimensional structure. Here we construct such a map using the recently collected corpora of high-throughput sequence data from the 75 base pairs long mutagenized E. coli lac promoter region, where each sequence is associated with its phenotype, the induced transcriptional activity measured by a fluorescent reporter. We find that the additive (non-epistatic) contributions of individual mutations account for about two-thirds of the explainable phenotype variance, while pairwise epistasis explains about 7% of the variance for the full mutagenized sequence and about 15% for the subsequence associated with protein binding sites. Surprisingly, there is no evidence for third order epistatic contributions, and our inferred fitness landscape is essentially single peaked, with a small amount of antagonistic epistasis. There is a significant selective pressure on the wild type, which we deduce to be multi-objective optimal for gene expression in environments with different nutrient sources. We identify transcription factor (CRP) and RNA polymerase binding sites in the promotor region and their interactions without difficult optimization steps. In particular, we observe evidence for previously unexplored genetic regulatory mechanisms, possibly kinetic in nature. We conclude with a cautionary note that inferred properties of fitness landscapes may be severely influenced by biases in the sequence data.

q-bio.PE↗

arXiv2012

Predictive information in a nonequilibrium critical model

We propose predictive information, that is information between a long past of duration T and the entire infinitely long future of a time series, as a universal order parameter to study phase transitions in physical systems. It can be used, in particular, to study nonequlibrium transitions and other exotic transitions, where a simpler order parameter cannot be identifies using traditional symmetry arguments. As an example, we calculate predictive information for a stochastic nonequilibrium dynamics problem that forms an absorbing state under a continuous change of a parameter. The information at the transition point diverges as log(T), and a smooth crossover to constant away from the transition is observed.

cond-mat.stat-mech↗

arXiv2012

Large number of receptors may reduce cellular response time variation

Cells often have tens of thousands of receptors, even though only a few activated receptors can trigger full cellular responses. Reasons for the overabundance of receptors remain unclear. We suggest that, in certain conditions, the large number of receptors results in a competition among receptors to be the first to activate the cell. The competition decreases the variability of the time to cellular activation, and hence results in a more synchronous activation of cells. We argue that, in simple models, this variability reduction does not necessarily interfere with the receptor specificity to ligands achieved by the kinetic proofreading mechanism. Thus cells can be activated accurately in time and specifically to certain signals. We predict the minimum number of receptors needed to reduce the coefficient of variation for the time to activation following binding of a specific ligand. Further, we predict the maximum number of receptors so that the kinetic proofreading mechanism still can improve the specificity of the activation. These predictions fall in line with experimentally reported receptor numbers for multiple systems.

q-bio.MN↗

arXiv2012

Population-expression models of immune response

The immune response to a pathogen has two basic features. The first is the expansion of a few pathogen-specific cells to form a population large enough to control the pathogen. The second is the process of differentiation of cells from an initial naive phenotype to an effector phenotype which controls the pathogen, and subsequently to a memory phenotype that is maintained and responsible for long-term protection. The expansion and the differentiation have been considered largely independently. Changes in cell populations are typically described using ecologically based ordinary differential equation models. In contrast, differentiation of single cells is studied within systems biology and is frequently modeled by considering changes in gene and protein expression in individual cells. Recent advances in experimental systems biology make available for the first time data to allow the coupling of population and high dimensional expression data of immune cells during infections. Here we describe and develop population-expression models which integrate these two processes into systems biology on the multicellular level. When translated into mathematical equations, these models result in non-conservative, non-local advection-diffusion equations. We describe situations where the population-expression approach can make correct inference from data while previous modeling approaches based on common simplifying assumptions would fail. We also explore how model reduction techniques can be used to build population-expression models, minimizing the complexity of the model while keeping the essential features of the system. While we consider problems in immunology in this paper, we expect population-expression models to be more broadly applicable.

q-bio.PE↗

arXiv2011

Gain control in molecular information processing: Lessons from neuroscience

Statistical properties of environments experienced by biological signaling systems in the real world change, which necessitate adaptive responses to achieve high fidelity information transmission. One form of such adaptive response is gain control. Here we argue that a certain simple mechanism of gain control, understood well in the context of systems neuroscience, also works for molecular signaling. The mechanism allows to transmit more than one bit (on or off) of information about the signal independently of the signal variance. It does not require additional molecular circuitry beyond that already present in many molecular systems, and, in particular, it does not depend on existence of feedback loops. The mechanism provides a potential explanation for abundance of ultrasensitive response curves in biological regulatory networks.

q-bio.MN↗

arXiv2011

Model Cortical Association Fields Account for the Time Course and Dependence on Target Complexity of Human Contour Perception

Can lateral connectivity in the primary visual cortex account for the time dependence and intrinsic task difficulty of human contour detection? To answer this question, we created a synthetic image set that prevents sole reliance on either low-level visual features or high-level context for the detection of target objects. Rendered images consist of smoothly varying, globally aligned contour fragments (amoebas) distributed among groups of randomly rotated fragments (clutter). The time course and accuracy of amoeba detection by humans was measured using a two-alternative forced choice protocol with self-reported confidence and variable image presentation time (20-200 ms), followed by an image mask optimized so as to interrupt visual processing. Measured psychometric functions were well fit by sigmoidal functions with exponential time constants of 30-91 ms, depending on amoeba complexity. Key aspects of the psychophysical experiments were accounted for by a computational network model, in which simulated responses across retinotopic arrays of orientation-selective elements were modulated by cortical association fields, represented as multiplicative kernels computed from the differences in pairwise edge statistics between target and distractor images. Comparing the experimental and the computational results suggests that each iteration of the lateral interactions takes at least 37.5 ms of cortical processing time. Our results provide evidence that cortical association fields between orientation selective elements in early visual areas can account for important temporal and task-dependent aspects of the psychometric curves characterizing human contour perception, with the remaining discrepancies postulated to arise from the influence of higher cortical areas.

q-bio.NC↗

arXiv2011

Fitness in time-dependent environments includes a geometric contribution

Phenotypic evolution implies sequential fixations of new genomic sequences. The speed at which these mutations fixate depends, in part, on the relative fitness (selection coefficient) of the mutant vs. the ancestor. Using a simple population dynamics model we show that the relative fitness in dynamical environments is not equal to the fitness averaged over individual environments. Instead it includes a term that explicitly depends on the sequence of the environments. This term is geometric in nature and depends only on the oriented area enclosed by the trajectory taken by the system in the environment state space. It is related to the well-studied geometric phases in classical and quantum physical systems. We discuss possible biological implications of these observations, focusing on evolution of novel metabolic or stress-resistant functions.

q-bio.PE↗

arXiv2010

Speeding up evolutionary search by small fitness fluctuations

We consider a fixed size population that undergoes an evolutionary adaptation in the weak mutuation rate limit, which we model as a biased Langevin process in the genotype space. We show analytically and numerically that, if the fitness landscape has a small highly epistatic (rough) and time-varying component, then the population genotype exhibits a high effective diffusion in the genotype space and is able to escape local fitness minima with a large probability. We argue that our principal finding that even very small time-dependent fluctuations of fitness can substantially speed up evolution is valid for a wide class of models.

q-bio.PE↗

arXiv2010

Information theory and adaptation

In this Chapter, we ask questions (1) What is the right way to measure the quality of information processing in a biological system? and (2) What can real-life organisms do in order to improve their performance in information-processing tasks? We then review the body of work that investigates these questions experimentally, computationally, and theoretically in biological domains as diverse as cell biology, population biology, and computational neuroscience

q-bio.QM↗

arXiv2010

Mass Conservation And Inference of Metabolic Networks from High-throughput Mass Spectrometry Data

We present a step towards the metabolome-wide computational inference of cellular metabolic reaction networks from metabolic profiling data, such as mass spectrometry. The reconstruction is based on identification of irreducible statistical interactions among the metabolite activities using the ARACNE reverse-engineering algorithm and on constraining possible metabolic transformations to satisfy the conservation of mass. The resulting algorithms are validated on synthetic data from an abridged computational model of Escherichia coli metabolism. Precision rates upwards of 50% are routinely observed for identification of full metabolic reactions, and recalls upwards of 20% are also seen.

q-bio.QM↗

arXiv2010

Multivariate dependence and genetic networks inference

A critical task in systems biology is the identification of genes that interact to control cellular processes by transcriptional activation of a set of target genes. Many methods have been developed to use statistical correlations in high-throughput datasets to infer such interactions. However, cellular pathways are highly cooperative, often requiring the joint effect of many molecules, and few methods have been proposed to explicitly identify such higher-order interactions, partially due to the fact that the notion of multivariate statistical dependency itself remains imprecisely defined. We define the concept of dependence among multiple variables using maximum entropy techniques and introduce computational tests for their identification. Synthetic network results reveal that this procedure uncovers dependencies even in undersampled regimes, when the joint probability distribution cannot be reliably estimated. Analysis of microarray data from human B cells reveals that third-order statistics, but not second-order ones, uncover relationships between genes that interact in a pathway to cooperatively regulate a common set of targets.

q-bio.QM↗

arXiv2009

Specificity and Completion Time Distributions of Biochemical Processes

In order to produce specific complex structures from a large set of similar biochemical building blocks, many biochemical systems require high sensitivity to small molecular differences. The first and most common model used to explain this high specificity is kinetic proofreading, which has been extended to a variety of systems from detection of DNA mismatch to cell signaling processes. While the specification properties of the kinetic proofreading model are well known and were studied in various contexts, very little is known about its temporal behavior. In this work, we study the dynamical properties of discrete stochastic two branch kinetic proofreading schemes. Using the Laplace transform of the corresponding chemical master equation, we obtain an analytical solution for the completion time distribution. In particular we provide expressions for the specificity and the mean and the variance of the process completion times. We also show that, for a wide range of parameters a process distinguishing between two different products can be reduced to a much simpler three point process. Our results allow for the systematic study of the interplay between specificity and completion times as well as testing the validity of the kinetic proofreading model in biological systems.

cond-mat.stat-mech↗

arXiv2009

Simplicity of Completion Time Distributions for Common Complex Biochemical Processes

Biochemical processes typically involve huge numbers of individual reversible steps, each with its own dynamical rate constants. For example, kinetic proofreading processes rely upon numerous sequential reactions in order to guarantee the precise construction of specific macromolecules. In this work, we study the transient properties of such systems and fully characterize their first passage (completion) time distributions. In particular, we provide explicit expressions for the mean and the variance of the completion time for a kinetic proofreading process and computational analyses for more complicated biochemical systems. We find that, for a wide range of parameters, as the system size grows, the completion time behavior simplifies: it becomes either deterministic or exponentially distributed, with a very narrow transition between the two regimes. In both regimes, the dynamical complexity of the full system is trivial compared to its apparent structural complexity. Similar simplicity is likely to arise in the dynamics of many complex multi-step biochemical processes. In particular, these findings suggest not only that one may not be able to understand individual elementary reactions from macroscopic observations, but also that such understanding may be unnecessary.

q-bio.MN↗

arXiv2009

Ergodic and Nonergodic Anomalous Diffusion in Coupled Stochastic Processes

Inspired by problems in biochemical kinetics, we study statistical properties of an overdamped Langevin process whose friction coefficient depends on the state of a similar, unobserved process. Integrating out the latter, we derive the long time behaviour of the mean square displacement. Anomalous diffusion is found. Since the diffusion exponent can not be predicted using a simple scaling argument, anomalous scaling appears as well. We also find that the coupling can lead to ergodic or non-ergodic behaviour of the studied process. We compare our theoretical predictions with numerical simulations and find an excellent agreement. The findings caution against treating biochemical systems coupled with unobserved dynamical degrees of freedom by means of standard, diffusive Langevin descriptions.

cond-mat.stat-mech↗

arXiv2008

Statistical properties of multistep enzyme-mediated reactions

Enzyme-mediated reactions may proceed through multiple intermediate conformational states before creating a final product molecule, and one often wishes to identify such intermediate structures from observations of the product creation. In this paper, we address this problem by solving the chemical master equations for various enzymatic reactions. We devise a perturbation theory analogous to that used in quantum mechanics that allows us to determine the first ( ) and the second (variance) cumulants of the distribution of created product molecules as a function of the substrate concentration and the kinetic rates of the intermediate processes. The mean product flux V=d /dt (or "dose-response" curve) and the Fano factor F=variance/ are both realistically measurable quantities, and while the mean flux can often appear the same for different reaction types, the Fano factor can be quite different. This suggests both qualitative and quantitative ways to discriminate between different reaction schemes, and we explore this possibility in the context of four sample multistep enzymatic reactions. We argue that measuring both the mean flux and the Fano factor can not only discriminate between reaction types, but can also provide some detailed information about the internal, unobserved kinetic rates, and this can be done without measuring single-molecule transition events.

q-bio.QM↗