Size matters: network inference tackles the genome scale

Transfer Learning for Molecular Cancer Classification Using Deep Neural Networks

The emergence of deep learning has impacted numerous machine learning based applications and research. The reason for its success lies in two main advantages: 1) it provides the ability to learn very complex non-linear relationships between features and 2) it allows one to leverage information from unlabeled data that does not belong to the problem being handled. This paper presents a transfer learning procedure for cancer classification, which uses feature selection and normalization techniques in conjunction with s sparse auto-encoders on gene expression data. While classifying any two tumor types, data of other tumor types were used in unsupervised manner to improve the feature representation. The performance of our algorithm was tested on 36 two-class benchmark datasets from the GEMLeR repository. On performing statistical tests, it is clearly ascertained that our algorithm statistically outperforms several generally used cancer classification approaches. The deep learning based molecular disease classification can be used to guide decisions made on the diagnosis and treatment of diseases, and therefore may have important applications in precision medicine.

Reverse engineering and identification in systems biology: strategies, perspectives and challenges

The interplay of mathematical modelling with experiments is one of the central elements in systems biology. The aim of reverse engineering is to infer, analyse and understand, through this interplay, the functional and regulatory mechanisms of biological systems. Reverse engineering is not exclusive of systems biology and has been studied in different areas, such as inverse problem theory, machine learning, nonlinear physics, (bio)chemical kinetics, control theory and optimization, among others. However, it seems that many of these areas have been relatively closed to outsiders. In this contribution, we aim to compare and highlight the different perspectives and contributions from these fields, with emphasis on two key questions: (i) why are reverse engineering problems so hard to solve, and (ii) what methods are available for the particular problems arising from systems biology?

Reverse engineering biomolecular systems using -omic data: challenges, progress and opportunities

Recent advances in high-throughput biotechnologies have led to the rapid growing research interest in reverse engineering of biomolecular systems (REBMS). 'Data-driven' approaches, i.e. data mining, can be used to extract patterns from large volumes of biochemical data at molecular-level resolution while 'design-driven' approaches, i.e. systems modeling, can be used to simulate emergent system properties. Consequently, both data- and design-driven approaches applied to -omic data may lead to novel insights in reverse engineering biological systems that could not be expected before using low-throughput platforms. However, there exist several challenges in this fast growing field of reverse engineering biomolecular systems: (i) to integrate heterogeneous biochemical data for data mining, (ii) to combine top-down and bottom-up approaches for systems modeling and (iii) to validate system models experimentally. In addition to reviewing progress made by the community and opportunities encountered in addressing these challenges, we explore the emerging field of synthetic biology, which is an exciting approach to validate and analyze theoretical system models directly through experimental synthesis, i.e. analysis-by-synthesis. The ultimate goal is to address the present and future challenges in reverse engineering biomolecular systems (REBMS) using integrated workflow of data mining, systems modeling and synthetic biology.

Generating stochastic gene regulatory networks consistent with pathway information and steady-state behavior

We present a procedure to generate a stochastic genetic regulatory network model consistent with pathway information. Using the stochastic dynamics of Markov chains, we produce a model constrained by the prior knowledge despite the sometimes incomplete, time independent, and often conflicting nature of these pathways. We apply the Markov theory to study the model's long run behavior and introduce a biologically important transformation to aid in comparison with real biological outcome prediction in the steady-state domain. Our technique produces biologically faithful models without the need for rate kinetics, detailed timing information, or complex inference procedures. To demonstrate the method, we produce a model using 28 pathways from the biological literature pertaining to the transcription factor family nuclear factor-κB. Predictions from this model in the steady-state domain are then validated against nine mice knockout experiments.

Predictive networks: a flexible, open source, web application for integration and analysis of human gene networks

Genomics provided us with an unprecedented quantity of data on the genes that are activated or repressed in a wide range of phenotypes. We have increasingly come to recognize that defining the networks and pathways underlying these phenotypes requires both the integration of multiple data types and the development of advanced computational methods to infer relationships between the genes and to estimate the predictive power of the networks through which they interact. To address these issues we have developed Predictive Networks (PN), a flexible, open-source, web-based application and data services framework that enables the integration, navigation, visualization and analysis of gene interaction networks. The primary goal of PN is to allow biomedical researchers to evaluate experimentally derived gene lists in the context of large-scale gene interaction networks. The PN analytical pipeline involves two key steps. The first is the collection of a comprehensive set of known gene interactions derived from a variety of publicly available sources. The second is to use these 'known' interactions together with gene expression data to infer robust gene networks. The PN web application is accessible from http://predictivenetworks.org. The PN code base is freely available at https://sourceforge.net/projects/predictivenets/.

Reverse engineering adverse outcome pathways

The toxicological effects of many stressors are mediated through unknown, or incompletely characterized, mechanisms of action. The application of reverse engineering complex interaction networks from high dimensional omics data (gene, protein, metabolic, signaling) can be used to overcome these limitations. This approach was used to characterize adverse outcome pathways (AOPs) for chemicals that disrupt the hypothalamus-pituitary-gonadal endocrine axis in fathead minnows (FHM, Pimephales promelas). Gene expression changes in FHM ovaries in response to seven different chemicals, over different times, doses, and in vivo versus in vitro conditions, were captured in a large data set of 868 arrays. Potential AOPs of the antiandrogen flutamide were examined using two mutual information-based methods to infer gene regulatory networks and potential AOPs. Representative networks from these studies were used to predict network paths from stressor to adverse outcome as candidate AOPs. The relationship of individual chemicals to an adverse outcome can be determined by following perturbations through the network in response to chemical treatment, thus leading to the nodes associated with the adverse outcome. Identification of candidate pathways allows for formation of testable hypotheses about key biological processes, biomarkers, or alternative endpoints that can be used to monitor an AOP. Finally, the unique challenges facing the application of this approach in ecotoxicology were identified and a road map for the utilization of these tools presented.

Adaptation of cells to new environments

The evolutionary success of an organism is a testament to its inherent capacity to keep pace with environmental conditions that change over short and long periods. Mechanisms underlying adaptive processes are being investigated with renewed interest and excitement. This revival is partly fueled by powerful technologies that can probe molecular phenomena at a systems scale. Such studies provide spectacular insight into the mechanisms of adaptation, including rewiring of regulatory networks via natural selection of horizontal gene transfers, gene duplication, deletion, readjustment of kinetic parameters, and myriad other genetic reorganizational events. Here, we discuss advances in prokaryotic systems biology from the perspective of evolutionary principles that have shaped regulatory networks for dynamic adaptation to environmental change.

DREAM3: network inference using dynamic context likelihood of relatedness and the inferelator

Background

Many current works aiming to learn regulatory networks from systems biology data must balance model complexity with respect to data availability and quality. Methods that learn regulatory associations based on unit-less metrics, such as Mutual Information, are attractive in that they scale well and reduce the number of free parameters (model complexity) per interaction to a minimum. In contrast, methods for learning regulatory networks based on explicit dynamical models are more complex and scale less gracefully, but are attractive as they may allow direct prediction of transcriptional dynamics and resolve the directionality of many regulatory interactions.

Methodology

We aim to investigate whether scalable information based methods (like the Context Likelihood of Relatedness method) and more explicit dynamical models (like Inferelator 1.0) prove synergistic when combined. We test a pipeline where a novel modification of the Context Likelihood of Relatedness (mixed-CLR, modified to use time series data) is first used to define likely regulatory interactions and then Inferelator 1.0 is used for final model selection and to build an explicit dynamical model.

Conclusions/Significance

Our method ranked 2nd out of 22 in the DREAM3 100-gene in silico networks challenge. Mixed-CLR and Inferelator 1.0 are complementary, demonstrating a large performance gain relative to any single tested method, with precision being especially high at low recall values. Partitioning the provided data set into four groups (knock-down, knock-out, time-series, and combined) revealed that using comprehensive knock-out data alone provides optimal performance. Inferelator 1.0 proved particularly powerful at resolving the directionality of regulatory interactions, i.e. “who regulates who” (approximately of identified true positives were correctly resolved). Performance drops for high in-degree genes, i.e. as the number of regulators per target gene increases, but not with out-degree, i.e. performance is not affected by the presence of regulatory hubs.

Inference of regulatory gene interactions from expression data using three-way mutual information

This paper describes the technique designated best performer in the 2nd conference on Dialogue for Reverse Engineering Assessments and Methods (DREAM2) Challenge 5 (unsigned genome-scale network prediction from blinded microarray data). Existing algorithms use the pairwise correlations of the expression levels of genes, which provide valuable but insufficient information for the inference of regulatory interactions. Here we present a computational approach based on the recently developed context likelihood of related (CLR) algorithm, extracting additional complementary information using the information theoretic measure of synergy and assigning a score to each ordered pair of genes measuring the degree of confidence that the first gene regulates the second. When tested on a set of publicly available Escherichia coli gene-expression data with known assumed ground truth, the synergy augmented CLR (SA-CLR) algorithm had significantly improved prediction performance when compared to CLR. There is also enhanced potential for biological discovery as a result of the identification of the most likely synergistic partner genes involved in the interactions.

A yeast synthetic network for in vivo assessment of reverse-engineering and modeling approaches

Systems biology approaches are extensively used to model and reverse engineer gene regulatory networks from experimental data. Conversely, synthetic biology allows "de novo" construction of a regulatory network to seed new functions in the cell. At present, the usefulness and predictive ability of modeling and reverse engineering cannot be assessed and compared rigorously. We built in the yeast Saccharomyces cerevisiae a synthetic network, IRMA, for in vivo "benchmarking" of reverse-engineering and modeling approaches. The network is composed of five genes regulating each other through a variety of regulatory interactions; it is negligibly affected by endogenous genes, and it is responsive to small molecules. We measured time series and steady-state expression data after multiple perturbations. These data were used to assess state-of-the-art modeling and reverse-engineering techniques. A semiquantitative model was able to capture and predict the behavior of the network. Reverse engineering based on differential equations and Bayesian networks correctly inferred regulatory interactions from the experimental data.

Biological Network Inference and analysis using SEBINI and CABIN

Attaining a detailed understanding of the various biological networks in an organism lies at the core of the emerging discipline of systems biology. A precise description of the relationships formed between genes, mRNA molecules, and proteins is a necessary step toward a complete description of the dynamic behavior of an organism at the cellular level, and toward intelligent, efficient, and directed modification of an organism. The importance of understanding such regulatory, signaling, and interaction networks has fueled the development of numerous in silico inference algorithms, as well as new experimental techniques and a growing collection of public databases. The Software Environment for BIological Network Inference (SEBINI) has been created to provide an interactive environment for the deployment, evaluation, and improvement of algorithms used to reconstruct the structure of biological regulatory and interaction networks. SEBINI can be used to analyze high-throughput gene expression, protein abundance, or protein activation data via a suite of state-of-the-art network inference algorithms. It also allows algorithm developers to compare and train network inference methods on artificial networks and simulated gene expression perturbation data. SEBINI can therefore be used by software developers wishing to evaluate, refine, or combine inference techniques, as well as by bioinformaticians analyzing experimental data. Networks inferred from the SEBINI software platform can be further analyzed using the Collective Analysis of Biological Interaction Networks (CABIN) tool, which is an exploratory data analysis software that enables integration and analysis of protein-protein interaction and gene-to-gene regulatory evidence obtained from multiple sources. The collection of edges in a public database, along with the confidence held in each edge (if available), can be fed into CABIN as one "evidence network," using the Cytoscape SIF file format. Using CABIN, one may increase the confidence in individual edges in a network inferred by an algorithm in SEBINI, as well as extend such a network by combining it with species-specific or generic information, e.g., known protein-protein interactions or target genes identified for known transcription factors. Thus, the combined SEBINI-CABIN toolkit aids in the more accurate reconstruction of biological networks, with less effort, in less time.A demonstration web site for SEBINI can be accessed from https://www.emsl.pnl.gov/SEBINI/RootServlet . Source code and PostgreSQL database schema are available under open source license.

CONTACT

ronald.taylor@pnl.gov. For commercial use, some algorithms included in SEBINI require licensing from the original developers. CABIN can be downloaded from http://www.sysbio.org/dataresources/cabin.stm .

CONTACT

mudita.singhal@pnl.gov.

Learning biological networks: from modules to dynamics

Learning regulatory networks from genomics data is an important problem with applications spanning all of biology and biomedicine. Functional genomics projects offer a cost-effective means of greatly expanding the completeness of our regulatory models, and for some prokaryotic organisms they offer a means of learning accurate models that incorporate the majority of the genome. There are, however, several reasons to believe that regulatory network inference is beyond our current reach, such as (i) the combinatorics of the problem, (ii) factors we can't (or don't often) collect genome-wide measurements for and (iii) dynamics that elude cost-effective experimental designs. Recent works have demonstrated the ability to reconstruct large fractions of prokaryotic regulatory networks from compendiums of genomics data; they have also demonstrated that these global regulatory models can be used to predict the dynamics of the transcriptome. We review an overall strategy for the reconstruction of global networks based on these results in microbial systems.

A predictive model for transcriptional control of physiology in a free living cell

The environment significantly influences the dynamic expression and assembly of all components encoded in the genome of an organism into functional biological networks. We have constructed a model for this process in Halobacterium salinarum NRC-1 through the data-driven discovery of regulatory and functional interrelationships among approximately 80% of its genes and key abiotic factors in its hypersaline environment. Using relative changes in 72 transcription factors and 9 environmental factors (EFs) this model accurately predicts dynamic transcriptional responses of all these genes in 147 newly collected experiments representing completely novel genetic backgrounds and environments-suggesting a remarkable degree of network completeness. Using this model we have constructed and tested hypotheses critical to this organism's interaction with its changing hypersaline environment. This study supports the claim that the high degree of connectivity within biological and EF networks will enable the construction of similar models for any organism from relatively modest numbers of experiments.

Systems biology and cancer stem cells

The identification, purification and characterization of cancer stem cells (CSCs) holds tremendous promise for improving the treatment of cancer. Mounting evidence is demonstrating that only certain tumour cells (i.e. the CSCs) can give rise to tumours when injected and that these purified cell populations generate heterogeneous tumours. While the cell of origin is still not determined definitively, specific molecular markers for populations containing these CSCs have been found for leukaemia, brain cancer and breast cancer, among others. Systems approaches, particularly molecular profiling, have proven to be of great utility for cancer diagnosis and characterization. These approaches also hold significant promise for identifying distinctive properties of the CSCs, and progress is already being made.

Scale relativity theory and integrative systems biology: 1. Founding principles and scale laws

In these two companion papers, we provide an overview and a brief history of the multiple roots, current developments and recent advances of integrative systems biology and identify multiscale integration as its grand challenge. Then we introduce the fundamental principles and the successive steps that have been followed in the construction of the scale relativity theory, and discuss how scale laws of increasing complexity can be used to model and understand the behaviour of complex biological systems. In scale relativity theory, the geometry of space is considered to be continuous but non-differentiable, therefore fractal (i.e., explicitly scale-dependent). One writes the equations of motion in such a space as geodesics equations, under the constraint of the principle of relativity of all scales in nature. To this purpose, covariant derivatives are constructed that implement the various effects of the non-differentiable and fractal geometry. In this first review paper, the scale laws that describe the new dependence on resolutions of physical quantities are obtained as solutions of differential equations acting in the scale space. This leads to several possible levels of description for these laws, from the simplest scale invariant laws to generalized laws with variable fractal dimensions. Initial applications of these laws to the study of species evolution, embryogenesis and cell confinement are discussed.