Non-linear reduction for kinetic models of metabolic reaction networks

Kinetic models of metabolic networks are essential for predicting and optimizing the transient behavior of cells in culture. However, such models are inherently high dimensional and stiff due to the large number of species and reactions involved and to kinetic rate constants of widely different orders of magnitude. In this paper we address the problem of deriving non-stiff, reduced-order non-linear models of the dominant dynamics of metabolic networks with fast and slow reactions. We present a method, based on singular perturbation analysis, which allows the systematic identification of quasi-steady-state conditions for the fast reactions, and the derivation of explicit non-linear models of the slow dynamics independent of the fast reaction rate expressions. The method is successfully applied to detailed models of metabolism in human erythrocytes and Saccharomyces cerevisiae.     

The dimensionality of ecological networks

How many dimensions (trait‐axes) are required to predict whether two species interact? This unanswered question originated with the idea of ecological niches, and yet bears relevance today for understanding what determines network structure. Here, we analyse a set of 200 ecological networks, including food webs, antagonistic and mutualistic networks, and find that the number of dimensions needed to completely explain all interactions is small ( < 10), with model selection favouring less than five. Using 18 high‐quality webs including several species traits, we identify which traits contribute the most to explaining network structure. We show that accounting for a few traits dramatically improves our understanding of the structure of ecological networks. Matching traits for resources and consumers, for example, fruit size and bill gape, are the most successful combinations. These results link ecologically important species attributes to large‐scale community structure.     

Bridging neuronal correlations and dimensionality reduction

1. Download : Download high-res image (119KB)
2. Download : Download full-size image

     

Haisu: Hierarchically supervised nonlinear dimensionality reduction

We propose a novel strategy for incorporating hierarchical supervised label information into nonlinear dimensionality reduction techniques. Specifically, we extend t-SNE, UMAP, and PHATE to include known or predicted class labels and demonstrate the efficacy of our approach on multiple single-cell RNA sequencing datasets. Our approach, “Haisu,” is applicable across domains and methods of nonlinear dimensionality reduction. In general, the mathematical effect of Haisu can be summarized as a variable perturbation of the high dimensional space in which the original data is observed. We thereby preserve the core characteristics of the visualization method and only change the manifold to respect known or assumed class labels when provided. Our strategy is designed to aid in the discovery and understanding of underlying patterns in a dataset that is heavily influenced by parent-child relationships. We show that using our approach can also help in semi-supervised settings where labels are known for only some datapoints (for instance when only a fraction of the cells are labeled). In summary, Haisu extends existing popular visualization methods to enable a user to incorporate labels known a priori into a visualization, including their hierarchical relationships as defined by a user input graph.     

Understanding protein flexibility through dimensionality reduction.

This work shows how to decrease the complexity of modeling flexibility in proteins by reducing the number of dimensions necessary to model important macromolecular motions such as the induced-fit process. Induced fit occurs during the binding of a protein to other proteins, nucleic acids, or small molecules (ligands) and is a critical part of protein function. It is now widely accepted that conformational changes of proteins can affect their ability to bind other molecules and that any progress in modeling protein motion and flexibility will contribute to the understanding of key biological functions. However, modeling protein flexibility has proven a very difficult task. Experimental laboratory methods, such as x-ray crystallography, produce rather limited information, while computational methods such as molecular dynamics are too slow for routine use with large systems. In this work, we show how to use the principal component analysis method, a dimensionality reduction technique, to transform the original high-dimensional representation of protein motion into a lower dimensional representation that captures the dominant modes of motions of proteins. For a medium-sized protein, this corresponds to reducing a problem with a few thousand degrees of freedom to one with less than fifty. Although there is inevitably some loss in accuracy, we show that we can obtain conformations that have been observed in laboratory experiments, starting from different initial conformations and working in a drastically reduced search space.     

Approaches to dimensionality reduction in proteomic biomarker studies

Mass-spectra based proteomic profiles have received widespreadattention as potential tools for biomarker discovery and earlydisease diagnosis. A major data-analytical problem involvedis the extremely high dimensionality (i.e. number of featuresor variables) of proteomic data, in particular when the samplesize is small. This article reviews dimensionality reductionmethods that have been used in proteomic biomarker studies.It then focuses on the problem of selecting the most appropriatemethod for a specific task or dataset, and proposes method combinationas a potential alternative to single-method selection. Finally,it points out the potential of novel dimension reduction techniques,in particular those that incorporate domain knowledge throughthe use of informative priors or causal inference.      

Rates of membrane-associated reactions: reduction of dimensionality revisited

The hypothesis that reactions associated with intracellular membranes enjoy a kinetic advantage from a reduced dimensionality for diffusion is inconsistent with available data on lateral diffusion rates, membrane-substrate affinities, and endogenous concentrations of enzymes and their aqueous substrates.     

A comparison of internal validation techniques for multifactor dimensionality reduction

Background  

It is hypothesized that common, complex diseases may be due to complex interactions between genetic and environmental factors, which are difficult to detect in high-dimensional data using traditional statistical approaches. Multifactor Dimensionality Reduction (MDR) is the most commonly used data-mining method to detect epistatic interactions. In all data-mining methods, it is important to consider internal validation procedures to obtain prediction estimates to prevent model over-fitting and reduce potential false positive findings. Currently, MDR utilizes cross-validation for internal validation. In this study, we incorporate the use of a three-way split (3WS) of the data in combination with a post-hoc pruning procedure as an alternative to cross-validation for internal model validation to reduce computation time without impairing performance. We compare the power to detect true disease causing loci using MDR with both 5- and 10-fold cross-validation to MDR with 3WS for a range of single-locus and epistatic disease models. Additionally, we analyze a dataset in HIV immunogenetics to demonstrate the results of the two strategies on real data.     

Modeling of signaling networks

Biochemical networks, including those containing signaling pathways, display a wide range of regulatory properties. These include the ability to propagate information across different time scales and to function as switches and oscillators. The mechanisms underlying these complex behaviors involve many interacting components and cannot be understood by experiments alone. The development of computational models and the integration of these models with experiments provide valuable insight into these complex systems-level behaviors. Here we review current approaches to the development of computational models of biochemical networks and describe the insights gained from models that integrate experimental data, using three examples that deal with ultrasensitivity, flexible bistability and oscillatory behavior. These types of complex behavior from relatively simple networks highlight the necessity of using theoretical approaches in understanding higher order biological functions.     

HillTau: A fast,compact abstraction for model reduction in biochemical signaling networks

Signaling networks mediate many aspects of cellular function. The conventional, mechanistically motivated approach to modeling such networks is through mass-action chemistry, which maps directly to biological entities and facilitates experimental tests and predictions. However such models are complex, need many parameters, and are computationally costly. Here we introduce the HillTau form for signaling models. HillTau retains the direct mapping to biological observables, but it uses far fewer parameters, and is 100 to over 1000 times faster than ODE-based methods. In the HillTau formalism, the steady-state concentration of signaling molecules is approximated by the Hill equation, and the dynamics by a time-course tau. We demonstrate its use in implementing several biochemical motifs, including association, inhibition, feedforward and feedback inhibition, bistability, oscillations, and a synaptic switch obeying the BCM rule. The major use-cases for HillTau are system abstraction, model reduction, scaffolds for data-driven optimization, and fast approximations to complex cellular signaling.     

Non-linear data structure extraction using simple hebbian networks

. We present a class a neural networks algorithms based on simple hebbian learning which allow the finding of higher order structure in data. The neural networks use negative feedback of activation to self-organise; such networks have previously been shown to be capable of performing principal component analysis (PCA). In this paper, this is extended to exploratory projection pursuit (EPP), which is a statistical method for investigating structure in high-dimensional data sets. As opposed to previous proposals for networks which learn using hebbian learning, no explicit weight normalisation, decay or weight clipping is required. The results are extended to multiple units and related to both the statistical literature on EPP and the neural network literature on non-linear PCA. Received: 30 May 1994/Accepted in revised form: 18 November 1994     

Lower-order effects adjustment in quantitative traits model-based multifactor dimensionality reduction

Identifying gene-gene interactions or gene-environment interactions in studies of human complex diseases remains a big challenge in genetic epidemiology. An additional challenge, often forgotten, is to account for important lower-order genetic effects. These may hamper the identification of genuine epistasis. If lower-order genetic effects contribute to the genetic variance of a trait, identified statistical interactions may simply be due to a signal boost of these effects. In this study, we restrict attention to quantitative traits and bi-allelic SNPs as genetic markers. Moreover, our interaction study focuses on 2-way SNP-SNP interactions. Via simulations, we assess the performance of different corrective measures for lower-order genetic effects in Model-Based Multifactor Dimensionality Reduction epistasis detection, using additive and co-dominant coding schemes. Performance is evaluated in terms of power and familywise error rate. Our simulations indicate that empirical power estimates are reduced with correction of lower-order effects, likewise familywise error rates. Easy-to-use automatic SNP selection procedures, SNP selection based on "top" findings, or SNP selection based on p-value criterion for interesting main effects result in reduced power but also almost zero false positive rates. Always accounting for main effects in the SNP-SNP pair under investigation during Model-Based Multifactor Dimensionality Reduction analysis adequately controls false positive epistasis findings. This is particularly true when adopting a co-dominant corrective coding scheme. In conclusion, automatic search procedures to identify lower-order effects to correct for during epistasis screening should be avoided. The same is true for procedures that adjust for lower-order effects prior to Model-Based Multifactor Dimensionality Reduction and involve using residuals as the new trait. We advocate using "on-the-fly" lower-order effects adjusting when screening for SNP-SNP interactions using Model-Based Multifactor Dimensionality Reduction analysis.     

Ovarian cancer classification based on dimensionality reduction for SELDI-TOF data

Background  

Recent advances in proteomics technologies such as SELDI-TOF mass spectrometry has shown promise in the detection of early stage cancers. However, dimensionality reduction and classification are considerable challenges in statistical machine learning. We therefore propose a novel approach for dimensionality reduction and tested it using published high-resolution SELDI-TOF data for ovarian cancer.     

Multifactor dimensionality reduction software for detecting gene-gene and gene-environment interactions

MOTIVATION: Polymorphisms in human genes are being described in remarkable numbers. Determining which polymorphisms and which environmental factors are associated with common, complex diseases has become a daunting task. This is partly because the effect of any single genetic variation will likely be dependent on other genetic variations (gene-gene interaction or epistasis) and environmental factors (gene-environment interaction). Detecting and characterizing interactions among multiple factors is both a statistical and a computational challenge. To address this problem, we have developed a multifactor dimensionality reduction (MDR) method for collapsing high-dimensional genetic data into a single dimension thus permitting interactions to be detected in relatively small sample sizes. In this paper, we describe the MDR approach and an MDR software package. RESULTS: We developed a program that integrates MDR with a cross-validation strategy for estimating the classification and prediction error of multifactor models. The software can be used to analyze interactions among 2-15 genetic and/or environmental factors. The dataset may contain up to 500 total variables and a maximum of 4000 study subjects. AVAILABILITY: Information on obtaining the executable code, example data, example analysis, and documentation is available upon request. SUPPLEMENTARY INFORMATION: All supplementary information can be found at http://phg.mc.vanderbilt.edu/Software/MDR.     

Modeling cell signaling networks

Cell signaling pathways interact with one another to form networks in mammalian systems. Such networks are complex in their organization and exhibit emergent properties such as bistability and ultrasensitivity. Analysis of signaling networks requires a combination of experimental and theoretical approaches including the development and analysis of models. This review focuses on theoretical approaches to understanding cell signaling networks. Using heterotrimeric G protein pathways an example, we demonstrate how interactions between two pathways can result in a network that contains a positive feedback loop and function as a switch. Different mathematical approaches that are currently used to model signaling networks are described, and future challenges including the need for databases as well as enhanced computing environments are discussed.     

Cell signaling networks ontology

Although databases for cell signaling pathways include numbers of reaction data of the pathways, the reaction data cannot be used yet to deduce biological functions from them. For the deduction, we need systematic and consistent interpretation of biological functions of reactions in cell signaling pathways in the context of "information transmission". To address this issue, we have developed a functional ontology for cell signaling pathways, Cell Signaling Network Ontology (CSN-Ontology), which provides framework for the functional interpretation presenting some important concepts as information, selectivity, movability, and signaling rules including passage of time.     

Quantitative analysis of signaling networks

The response of biological cells to environmental change is coordinated by protein-based signaling networks. These networks are to be found in both prokaryotes and eukaryotes. In eukaryotes, the signaling networks can be highly complex, some networks comprising of 60 or more proteins. The fundamental motif that has been found in all signaling networks is the protein phosphorylation/dephosphorylation cycle--the cascade cycle. At this time, the computational function of many of the signaling networks is poorly understood. However, it is clear that it is possible to construct a huge variety of control and computational circuits, both analog and digital from combinations of the cascade cycle. In this review, we will summarize the great versatility of the simple cascade cycle as a computational unit and towards the end give two examples, one prokaryotic chemotaxis circuit and the other, the eukaryotic MAPK cascade.