Chemogenomic profiling on a genome-wide scale using reverse-engineered gene networks

Drug target prediction through deep learning functional representation of gene signatures

Many machine learning applications in bioinformatics currently rely on matching gene identities when analyzing input gene signatures and fail to take advantage of preexisting knowledge about gene functions. To further enable comparative analysis of OMICS datasets, including target deconvolution and mechanism of action studies, we develop an approach that represents gene signatures projected onto their biological functions, instead of their identities, similar to how the word2vec technique works in natural language processing. We develop the Functional Representation of Gene Signatures (FRoGS) approach by training a deep learning model and demonstrate that its application to the Broad Institute's L1000 datasets results in more effective compound-target predictions than models based on gene identities alone. By integrating additional pharmacological activity data sources, FRoGS significantly increases the number of high-quality compound-target predictions relative to existing approaches, many of which are supported by in silico and/or experimental evidence. These results underscore the general utility of FRoGS in machine learning-based bioinformatics applications. Prediction networks pre-equipped with the knowledge of gene functions may help uncover new relationships among gene signatures acquired by large-scale OMICs studies on compounds, cell types, disease models, and patient cohorts.

Selenium-More than Just a Fortuitous Sulfur Substitute in Redox Biology

Living organisms use selenium mainly in the form of selenocysteine in the active site of oxidoreductases. Here, selenium's unique chemistry is believed to modulate the reaction mechanism and enhance the catalytic efficiency of specific enzymes in ways not achievable with a sulfur-containing cysteine. However, despite the fact that selenium/sulfur have different physicochemical properties, several selenoproteins have fully functional cysteine-containing homologues and some organisms do not use selenocysteine at all. In this review, selected selenocysteine-containing proteins will be discussed to showcase both situations: (i) selenium as an obligatory element for the protein's physiological function, and (ii) selenium presenting no clear advantage over sulfur (functional proteins with either selenium or sulfur). Selenium's physiological roles in antioxidant defence (to maintain cellular redox status/hinder oxidative stress), hormone metabolism, DNA synthesis, and repair (maintain genetic stability) will be also highlighted, as well as selenium's role in human health. Formate dehydrogenases, hydrogenases, glutathione peroxidases, thioredoxin reductases, and iodothyronine deiodinases will be herein featured.

A framework for considering prior information in network-based approaches to omics data analysis

For decades, molecular biologists have been uncovering the mechanics of biological systems. Efforts to bring their findings together have led to the development of multiple databases and information systems that capture and present pathway information in a computable network format. Concurrently, the advent of modern omics technologies has empowered researchers to systematically profile cellular processes across different modalities. Numerous algorithms, methodologies, and tools have been developed to use prior knowledge networks (PKNs) in the analysis of omics datasets. Interestingly, it has been repeatedly demonstrated that the source of prior knowledge can greatly impact the results of a given analysis. For these methods to be successful it is paramount that their selection of PKNs is amenable to the data type and the computational task they aim to accomplish. Here we present a five-level framework that broadly describes network models in terms of their scope, level of detail, and ability to inform causal predictions. To contextualize this framework, we review a handful of network-based omics analysis methods at each level, while also describing the computational tasks they aim to accomplish.

Learning perturbation-inducible cell states from observability analysis of transcriptome dynamics

A major challenge in biotechnology and biomanufacturing is the identification of a set of biomarkers for perturbations and metabolites of interest. Here, we develop a data-driven, transcriptome-wide approach to rank perturbation-inducible genes from time-series RNA sequencing data for the discovery of analyte-responsive promoters. This provides a set of biomarkers that act as a proxy for the transcriptional state referred to as cell state. We construct low-dimensional models of gene expression dynamics and rank genes by their ability to capture the perturbation-specific cell state using a novel observability analysis. Using this ranking, we extract 15 analyte-responsive promoters for the organophosphate malathion in the underutilized host organism Pseudomonas fluorescens SBW25. We develop synthetic genetic reporters from each analyte-responsive promoter and characterize their response to malathion. Furthermore, we enhance malathion reporting through the aggregation of the response of individual reporters with a synthetic consortium approach, and we exemplify the library's ability to be useful outside the lab by detecting malathion in the environment. The engineered host cell, a living malathion sensor, can be optimized for use in environmental diagnostics while the developed machine learning tool can be applied to discover perturbation-inducible gene expression systems in the compendium of host organisms.

Expanding the search for small-molecule antibacterials by multidimensional profiling

New techniques for systematic profiling of small-molecule effects can enhance traditional growth inhibition screens for antibiotic discovery and change how we search for new antibacterial agents. Computational models that integrate physicochemical compound properties with their phenotypic and molecular downstream effects can not only predict efficacy of molecules yet to be tested, but also reveal unprecedented insights on compound modes of action (MoAs). The unbiased characterization of compounds that themselves are not growth inhibitory but exhibit diverse MoAs, can expand antibacterial strategies beyond direct inhibition of core essential functions. Early and systematic functional annotation of compound libraries thus paves the way to new models in the selection of lead antimicrobial compounds. In this Review, we discuss how multidimensional small-molecule profiling and the ever-increasing computing power are accelerating the discovery of unconventional antibacterials capable of bypassing resistance and exploiting synergies with established antibacterial treatments and with protective host mechanisms.

Computational Framework for Machine-Learning-Enabled 13C Fluxomics

¹³C metabolic flux analysis (MFA) has emerged as a powerful tool for synthetic biology. This optimization-based approach suffers long computation time and unstable solutions depending on the initial guess. Here, we develop a machine-learning-based framework for ¹³C fluxomics. Specifically, training and test data sets are generated by metabolic network decomposition and flux sampling, in which flux ratios at metabolic nodes and simulated labeling patterns of metabolites are used as training targets and features, respectively. To improve prediction accuracy and simplify the model, automated processes are developed for flux ratio selection based on solvability and feature screening based on importance. We found that predictive performance can be significantly improved using both amino acids and central carbon metabolites in comparison with amino acids alone. Together with measured external fluxes, the predicted flux ratios determine the mass balance system, yielding global flux distributions. This approach is validated by flux estimation using both simulated and experimental data in comparison with canonical ¹³C MFA. The approach represents a reliable fluxomics method readily applicable to high-throughput metabolic phenotyping, which highlights the advances of intelligent learning algorithms in synthetic biology, specifically in the Test and Learn stage of the Design-Build-Test-Learn cycle.

Reconstruction of dynamic regulatory networks reveals signaling-induced topology changes associated with germ layer specification

Elucidating regulatory relationships between transcription factors (TFs) and target genes is fundamental to understanding how cells control their identity and behavior. Unfortunately, existing computational gene regulatory network (GRN) reconstruction methods are imprecise, computationally burdensome, and fail to reveal dynamic regulatory topologies. Here, we present Epoch, a reconstruction tool that uses single-cell transcriptomics to accurately infer dynamic networks. We apply Epoch to identify the dynamic networks underpinning directed differentiation of mouse embryonic stem cells (ESCs) guided by multiple signaling pathways, and we demonstrate that modulating these pathways drives topological changes that bias cell fate potential. We also find that Peg3 rewires the pluripotency network to favor mesoderm specification. By integrating signaling pathways with GRNs, we trace how Wnt activation and PI3K suppression govern mesoderm and endoderm specification, respectively. Finally, we identify regulatory circuits of patterning and axis formation that distinguish in vitro and in vivo mesoderm specification.

Drug target inference by mining transcriptional data using a novel graph convolutional network framework

A fundamental challenge that arises in biomedicine is the need to characterize compounds in a relevant cellular context in order to reveal potential on-target or off-target effects. Recently, the fast accumulation of gene transcriptional profiling data provides us an unprecedented opportunity to explore the protein targets of chemical compounds from the perspective of cell transcriptomics and RNA biology. Here, we propose a novel Siamese spectral-based graph convolutional network (SSGCN) model for inferring the protein targets of chemical compounds from gene transcriptional profiles. Although the gene signature of a compound perturbation only provides indirect clues of the interacting targets, and the biological networks under different experiment conditions further complicate the situation, the SSGCN model was successfully trained to learn from known compound-target pairs by uncovering the hidden correlations between compound perturbation profiles and gene knockdown profiles. On a benchmark set and a large time-split validation dataset, the model achieved higher target inference accuracy as compared to previous methods such as Connectivity Map. Further experimental validations of prediction results highlight the practical usefulness of SSGCN in either inferring the interacting targets of compound, or reversely, in finding novel inhibitors of a given target of interest.

DeltaNeTS+: elucidating the mechanism of drugs and diseases using gene expression and transcriptional regulatory networks

BACKGROUND

Knowledge on the molecular targets of diseases and drugs is crucial for elucidating disease pathogenesis and mechanism of action of drugs, and for driving drug discovery and treatment formulation. In this regard, high-throughput gene transcriptional profiling has become a leading technology, generating whole-genome data on the transcriptional alterations caused by diseases or drug compounds. However, identifying direct gene targets, especially in the background of indirect (downstream) effects, based on differential gene expressions is difficult due to the complexity of gene regulatory network governing the gene transcriptional processes.

RESULTS

In this work, we developed a network analysis method, called DeltaNeTS+, for inferring direct gene targets of drugs and diseases from gene transcriptional profiles. DeltaNeTS+ uses a gene regulatory network model to identify direct perturbations to the transcription of genes using gene expression data. Importantly, DeltaNeTS+ is able to combine both steady-state and time-course expression profiles, as well as leverage information on the gene network structure. We demonstrated the power of DeltaNeTS+ in predicting gene targets using gene expression data in complex organisms, including Caenorhabditis elegans and human cell lines (T-cell and Calu-3). More specifically, in an application to time-course gene expression profiles of influenza A H1N1 (swine flu) and H5N1 (avian flu) infection, DeltaNeTS+ shed light on the key differences of dynamic cellular perturbations caused by the two influenza strains.

CONCLUSION

DeltaNeTS+ is a powerful network analysis tool for inferring gene targets from gene expression profiles. As demonstrated in the case studies, by incorporating available information on gene network structure, DeltaNeTS+ produces accurate predictions of direct gene targets from a small sample size (~ 10 s). Integrating static and dynamic expression data with transcriptional network structure extracted from genomic information, as enabled by DeltaNeTS+, is crucial toward personalized medicine, where treatments can be tailored to individual patients. DeltaNeTS+ can be freely downloaded from http://www.github.com/cabsel/deltanetsplus .

TREAP: A New Topological Approach to Drug Target Inference

Over 50% of drugs fail in stage 3 clinical trials, many because of a poor understanding of the drug's mechanisms of action (MoA). A better comprehension of drug MoA will significantly improve research and development (R&D). Current proposed algorithms, such as ProTINA and DeMAND, can be overly complex. Additionally, they are unable to predict whether the drug-induced gene expression or the topology of the networks used to model gene regulation primarily impacts accurate drug target inference. In this work, we evaluate how network and gene expression data affect ProTINA's accuracy. We find that network topology predominantly determines the accuracy of ProTINA's predictions. We further show that the size of an interaction network and/or selecting cell-specific networks has a limited effect on accuracy. We then demonstrate that a specific network topology measure, betweenness, can be used to improve drug target prediction. Based on these results, we create a new algorithm, TREAP, that combines betweenness values and adjusted p-values for target inference. TREAP offers an alternative approach to drug target inference and is advantageous because it is not computationally demanding, provides easy-to-interpret results, and is often more accurate at predicting drug targets than current state-of-the-art approaches.

DOKI: Domain knowledge-driven inference method for reverse-engineering transcriptional regulatory relationships among genes in cancer

Efficient reverse-engineering methods are important for identifying transcriptional regulatory relationships among genes in cancer. These methods are becoming increasingly useful in this era where huge volumes of data are generated through the use of high-throughput technologies such as next-generation sequencing technologies and microarrays. However, it is important to improve current methods because of complications involved in modelling complex biological systems. In this paper, we present a novel approach, Domain Knowledge-driven Inference (DOKI), for identification of transcriptional regulatory relationships among genes, given a biological context such as cancer. Combining data normalization, the use of a probability distribution function and Kullback-Leibler Divergence, DOKI incorporates a domain knowledge-driven criterion to make determinations of the existence of regulatory relationships between given transcription factors and given specific gene targets. Characteristics of DOKI enable it to adequately handle complexities inherent in data, and accurately unearth linear and higher-order dependent relationships among genes. DOKI performed equally well with one established high-performing method and better than three other high-performing methods on relatively small data sets. However, it remarkably outperformed these methods on larger data sets to demonstrate its utility. Furthermore, we demonstrate the relevance of such inference algorithms for identifying novel relationships among genes in breast cancer, as some of the consensus results representing novel relationships were confirmed in previously published experimental results. Thus, DOKI will facilitate current efforts to gain etiological insights and help uncover new targeted therapies for various diseases.

Entropy of a bacterial stress response is a generalizable predictor for fitness and antibiotic sensitivity

Current approaches explore bacterial genes that change transcriptionally upon stress exposure as diagnostics to predict antibiotic sensitivity. However, transcriptional changes are often specific to a species or antibiotic, limiting implementation to known settings only. While a generalizable approach, predicting bacterial fitness independent of strain, species or type of stress, would eliminate such limitations, it is unclear whether a stress-response can be universally captured. By generating a multi-stress and species RNA-Seq and experimental evolution dataset, we highlight the strengths and limitations of existing gene-panel based methods. Subsequently, we build a generalizable method around the observation that global transcriptional disorder seems to be a common, low-fitness, stress response. We quantify this disorder using entropy, which is a specific measure of randomness, and find that in low fitness cases increasing entropy and transcriptional disorder results from a loss of regulatory gene-dependencies. Using entropy as a single feature, we show that fitness and quantitative antibiotic sensitivity predictions can be made that generalize well beyond training data. Furthermore, we validate entropy-based predictions in 7 species under antibiotic and non-antibiotic conditions. By demonstrating the feasibility of universal predictions of bacterial fitness, this work establishes the fundamentals for potentially new approaches in infectious disease diagnostics.

Conceptual and computational framework for logical modelling of biological networks deregulated in diseases

Mathematical models can serve as a tool to formalize biological knowledge from diverse sources, to investigate biological questions in a formal way, to test experimental hypotheses, to predict the effect of perturbations and to identify underlying mechanisms. We present a pipeline of computational tools that performs a series of analyses to explore a logical model's properties. A logical model of initiation of the metastatic process in cancer is used as a transversal example. We start by analysing the structure of the interaction network constructed from the literature or existing databases. Next, we show how to translate this network into a mathematical object, specifically a logical model, and how robustness analyses can be applied to it. We explore the visualization of the stable states, defined as specific attractors of the model, and match them to cellular fates or biological read-outs. With the different tools we present here, we explain how to assign to each solution of the model a probability and how to identify genetic interactions using mutant phenotype probabilities. Finally, we connect the model to relevant experimental data: we present how some data analyses can direct the construction of the network, and how the solutions of a mathematical model can also be compared with experimental data, with a particular focus on high-throughput data in cancer biology. A step-by-step tutorial is provided as a Supplementary Material and all models, tools and scripts are provided on an accompanying website: https://github.com/sysbio-curie/Logical_modelling_pipeline.

Reverse engineering gene networks using global-local shrinkage rules

Inferring gene regulatory networks from high-throughput 'omics' data has proven to be a computationally demanding task of critical importance. Frequently, the classical methods break down owing to the curse of dimensionality, and popular strategies to overcome this are typically based on regularized versions of the classical methods. However, these approaches rely on loss functions that may not be robust and usually do not allow for the incorporation of prior information in a straightforward way. Fully Bayesian methods are equipped to handle both of these shortcomings quite naturally, and they offer the potential for improvements in network structure learning. We propose a Bayesian hierarchical model to reconstruct gene regulatory networks from time-series gene expression data, such as those common in perturbation experiments of biological systems. The proposed methodology uses global-local shrinkage priors for posterior selection of regulatory edges and relaxes the common normal likelihood assumption in order to allow for heavy-tailed data, which were shown in several of the cited references to severely impact network inference. We provide a sufficient condition for posterior propriety and derive an efficient Markov chain Monte Carlo via Gibbs sampling in the electronic supplementary material. We describe a novel way to detect multiple scales based on the corresponding posterior quantities. Finally, we demonstrate the performance of our approach in a simulation study and compare it with existing methods on real data from a T-cell activation study.

Integration of Multiple Data Sources for Gene Network Inference Using Genetic Perturbation Data

The inference of gene networks from large-scale human genomic data is challenging due to the difficulty in identifying correct regulators for each gene in a high-dimensional search space. We present a Bayesian approach integrating external data sources with knockdown data from human cell lines to infer gene regulatory networks. In particular, we assemble multiple data sources, including gene expression data, genome-wide binding data, gene ontology, and known pathways, and use a supervised learning framework to compute prior probabilities of regulatory relationships. We show that our integrated method improves the accuracy of inferred gene networks as well as extends some previous Bayesian frameworks both in theory and applications. We apply our method to two different human cell lines, namely skin melanoma cell line A375 and lung cancer cell line A549, to illustrate the capabilities of our method. Our results show that the improvement in performance could vary from cell line to cell line and that we might need to choose different external data sources serving as prior knowledge if we hope to obtain better accuracy for different cell lines.

Network perturbation analysis of gene transcriptional profiles reveals protein targets and mechanism of action of drugs and influenza A viral infection

Genome-wide transcriptional profiling provides a global view of cellular state and how this state changes under different treatments (e.g. drugs) or conditions (e.g. healthy and diseased). Here, we present ProTINA (Protein Target Inference by Network Analysis), a network perturbation analysis method for inferring protein targets of compounds from gene transcriptional profiles. ProTINA uses a dynamic model of the cell-type specific protein-gene transcriptional regulation to infer network perturbations from steady state and time-series differential gene expression profiles. A candidate protein target is scored based on the gene network's dysregulation, including enhancement and attenuation of transcriptional regulatory activity of the protein on its downstream genes, caused by drug treatments. For benchmark datasets from three drug treatment studies, ProTINA was able to provide highly accurate protein target predictions and to reveal the mechanism of action of compounds with high sensitivity and specificity. Further, an application of ProTINA to gene expression profiles of influenza A viral infection led to new insights of the early events in the infection.

Efficient proximal gradient algorithm for inference of differential gene networks

Background

Gene networks in living cells can change depending on various conditions such as caused by different environments, tissue types, disease states, and development stages. Identifying the differential changes in gene networks is very important to understand molecular basis of various biological process. While existing algorithms can be used to infer two gene networks separately from gene expression data under two different conditions, and then to identify network changes, such an approach does not exploit the similarity between two gene networks, and it is thus suboptimal. A desirable approach would be clearly to infer two gene networks jointly, which can yield improved estimates of network changes.

Results

In this paper, we developed a proximal gradient algorithm for differential network (ProGAdNet) inference, that jointly infers two gene networks under different conditions and then identifies changes in the network structure. Computer simulations demonstrated that our ProGAdNet outperformed existing algorithms in terms of inference accuracy, and was much faster than a similar approach for joint inference of gene networks. Gene expression data of breast tumors and normal tissues in the TCGA database were analyzed with our ProGAdNet, and revealed that 268 genes were involved in the changed network edges. Gene set enrichment analysis identified a significant number of gene sets related to breast cancer or other types of cancer that are enriched in this set of 268 genes. Network analysis of the kidney cancer data in the TCGA database with ProGAdNet also identified a set of genes involved in network changes, and the majority of the top genes identified have been reported in the literature to be implicated in kidney cancer. These results corroborated that the gene sets identified by ProGAdNet were very informative about the cancer disease status. A software package implementing the ProGAdNet, computer simulations, and real data analysis is available as Additional file 1.

Conclusion

With its superior performance over existing algorithms, ProGAdNet provides a valuable tool for finding changes in gene networks, which may aid the discovery of gene-gene interactions changed under different conditions.

Electronic supplementary material

The online version of this article (10.1186/s12859-019-2749-x) contains supplementary material, which is available to authorized users.

Modelling and Control of Gene Regulatory Networks for Perturbation Mitigation

Synthetic Biologists are increasingly interested in the idea of using synthetic feedback control circuits for the mitigation of perturbations to gene regulatory networks that may arise due to disease and/or environmental disturbances. Models employing Michaelis-Menten kinetics with Hill-type nonlinearities are typically used to represent the dynamics of gene regulatory networks. Here, we identify some fundamental problems with such models from the point of view of control system design, and argue that an alternative formalism, based on so-called S-System models, is more suitable. Using tools from system identification, we show how to build S-System models that capture the key dynamics of an example gene regulatory network, and design a genetic feedback controller with the objective of rejecting an external perturbation. Using a sine sweeping method, we show how the S-System model can be approximated by a linear transfer function and, based on this transfer function, we design our controller. Simulation results using the full nonlinear S-System model of the network show that the synthetic control circuit is able to mitigate the effect of external perturbations. Our study is the first to highlight the usefulness of the S-System modelling formalism for the design of synthetic control circuits for gene regulatory networks.

Restricted-derestricted dynamic Bayesian Network inference of transcriptional regulatory relationships among genes in cancer

Understanding transcriptional regulatory relationships among genes is important for gaining etiological insights into diseases such as cancer. To this end, high-throughput biological data have been generated through advancements in a variety of technologies. These rely on computational approaches to discover underlying structures in such data. Among these computational approaches, Bayesian networks (BNs) stand out because their probabilistic nature enables them to manage randomness in the dynamics of gene regulation and experimental data. Feedback loops inherent in networks of regulatory relationships are more tractable when enhancements to BNs are applied to them. Here, we propose Restricted-Derestricted dynamic BNs with a novel search technique, Restricted-Derestricted Greedy Method, for such tasks. This approach relies on the Restricted-Derestricted Greedy search technique to infer transcriptional regulatory networks in two phases: restricted inference and derestricted inference. An application of this approach to real data sets reveals it performs favourably well compared to other existing well performing dynamic BN approaches in terms of recovering true relationships among genes. In addition, it provides a balance between searching for optimal networks and keeping biologically relevant regulatory interactions among variables.

Ambiguity in logic-based models of gene regulatory networks: An integrative multi-perturbation analysis

Most studies of gene regulatory network (GRN) inference have focused extensively on identifying the interaction map of the GRNs. However, in order to predict the cellular behavior, modeling the GRN in terms of logic circuits, i.e., Boolean networks, is necessary. The perturbation techniques, e.g., knock-down and over-expression, should be utilized for identifying the underlying logic behind the interactions. However, we will show that by using only transcriptomic data obtained by single-perturbation experiments, we cannot observe all regulatory interactions, and this invisibility causes ambiguity in our model. Consequently, we need to employ the data of multiple omics layers (genome, transcriptome, and proteome) as well as multiple perturbation experiments to reduce or eliminate ambiguity in our modeling. In this paper, we introduce a multi-step perturbation experiment to deal with ambiguity. Moreover, we perform a thorough analysis to investigate which types of perturbations and omics layers play the most important role in the unambiguous modeling of the GRNs and how much ambiguity will be eliminated by considering more perturbations and more omics layers. Our analysis shows that performing both knock-down and over-expression is necessary in order to achieve the least ambiguous model. Moreover, the more steps of the perturbation are taken, the more ambiguity is eliminated. In addition, we can even achieve an unambiguous model of the GRN by using multi-step perturbation and integrating transcriptomic, protein-protein interaction, and cis-element data. Finally, we demonstrate the effect of utilizing different types of perturbation experiment and integrating multi-omics data on identifying the logic behind the regulatory interactions in a synthetic GRN. In conclusion, relying on the results of only knock-down experiments and not including as many omics layers as possible in the GRN inference, makes the results ambiguous, unreliable, and less accurate.

Regulation of Adipogenesis and Thermogenesis through Mouse Olfactory Receptor 23 Stimulated by α-Cedrene in 3T3-L1 Cells

Olfactory receptors (ORs) are G protein-coupled receptors that perform important physiological functions beyond their role as odorant detectors in the olfactory sensory neurons. In the present study, we describe a novel role for one of these ORs, mouse olfactory receptor 23 (MOR23), as a regulator of adipogenesis and thermogenesis in 3T3-L1 cells. Downregulation of MOR23 by small interfering RNA in 3T3-L1 cells enhanced intracellular lipid accumulation and reduced the oxygen consumption rate. In agreement with this phenotype, MOR23 deletion significantly decreased intracellular cyclic adenosine monophosphate (cAMP) levels and protein amounts of adenylyl cyclase 3 (ADCY3), protein kinase A catalytic subunit (PKA Cα), phospho-5′-adenosine monophosphate (AMP)-activated protein kinase (AMPK), and phospho-cAMP-responsive element-binding protein (CREB), along with upregulation of adipogenic genes and downregulation of genes involved in thermogenesis. Activation of MOR23 by α-cedrene, a novel natural ligand of MOR23, significantly reduced lipid content, increased the oxygen consumption rate, and stimulated reprogramming of the metabolic signature of 3T3-L1 cells, and these changes elicited by α-cedrene were absent in MOR23-deficient cells. These findings point to the role of MOR23 as a regulator of adipogenesis and thermogenesis in adipocytes.

Parallel Algorithms for Inferring Gene Regulatory Networks: A Review

System biology problems such as whole-genome network construction from large-scale gene expression data are sophisticated and time-consuming. Therefore, using sequential algorithms are not feasible to obtain a solution in an acceptable amount of time. Today, by using massively parallel computing, it is possible to infer large-scale gene regulatory networks. Recently, establishing gene regulatory networks from large-scale datasets have drawn the noticeable attention of researchers in the field of parallel computing and system biology. In this paper, we attempt to provide a more detailed overview of the recent parallel algorithms for constructing gene regulatory networks. Firstly, fundamentals of gene regulatory networks inference and large-scale datasets challenges are given. Secondly, a detailed description of the four parallel frameworks and libraries including CUDA, OpenMP, MPI, and Hadoop is discussed. Thirdly, parallel algorithms are reviewed. Finally, some conclusions and guidelines for parallel reverse engineering are described.

A Framework for Engineering Stress Resilient Plants Using Genetic Feedback Control and Regulatory Network Rewiring

Crop disease leads to significant waste worldwide, both pre- and postharvest, with subsequent economic and sustainability consequences. Disease outcome is determined both by the plants' response to the pathogen and by the ability of the pathogen to suppress defense responses and manipulate the plant to enhance colonization. The defense response of a plant is characterized by significant transcriptional reprogramming mediated by underlying gene regulatory networks, and components of these networks are often targeted by attacking pathogens. Here, using gene expression data from Botrytis cinerea-infected Arabidopsis plants, we develop a systematic approach for mitigating the effects of pathogen-induced network perturbations, using the tools of synthetic biology. We employ network inference and system identification techniques to build an accurate model of an Arabidopsis defense subnetwork that contains key genes determining susceptibility of the plant to the pathogen attack. Once validated against time-series data, we use this model to design and test perturbation mitigation strategies based on the use of genetic feedback control. We show how a synthetic feedback controller can be designed to attenuate the effect of external perturbations on the transcription factor CHE in our subnetwork. We investigate and compare two approaches for implementing such a controller biologically-direct implementation of the genetic feedback controller, and rewiring the regulatory regions of multiple genes-to achieve the network motif required to implement the controller. Our results highlight the potential of combining feedback control theory with synthetic biology for engineering plants with enhanced resilience to environmental stress.

Next-Generation Machine Learning for Biological Networks

Machine learning, a collection of data-analytical techniques aimed at building predictive models from multi-dimensional datasets, is becoming integral to modern biological research. By enabling one to generate models that learn from large datasets and make predictions on likely outcomes, machine learning can be used to study complex cellular systems such as biological networks. Here, we provide a primer on machine learning for life scientists, including an introduction to deep learning. We discuss opportunities and challenges at the intersection of machine learning and network biology, which could impact disease biology, drug discovery, microbiome research, and synthetic biology.

Chemical genomic analysis of GPR35 signaling

GPR35, a family A orphan G protein-coupled receptor, has been implicated in inflammatory, neurological, and cardiovascular diseases. However, not much is known about the signaling and functions of GPR35. We performed a label-free kinome short hairpin RNA screen and identified a putative signaling network of GPR35 in HT-29 cells, some of which was validated using gene expression, biochemical and cellular assays. The results showed that GPR35 induced hypoxia-inducible factor 1α, and was involved in synaptic transmission, sensory perception, the immune system, and morphogenetic processes. Collectively, our data suggest that GPR35 may play an important role in response to hypoxic stress and be a potential target for the treatment of inflammatory, cardiovascular, and neurological disorders.

Olfactory receptor 10J5 responding to α-cedrene regulates hepatic steatosis via the cAMP-PKA pathway

Ectopic expression and functions of odorant receptors (ORs) in the human body have aroused much interest in the past decade. Mouse olfactory receptor 23 (MOR23, olfr16) and its human orthologue, OR10J5, have been found to be functionally expressed in several non-olfactory systems. Here, using MOR23- and OR10J5-expressing Hana3A cells, we identified α-cedrene, a natural compound that protects against hepatic steatosis in mice fed the high-fat diet, as a novel agonist of these receptors. In human hepatocytes, an RNA interference-mediated knockdown of OR10J5 increased intracellular lipid accumulation, along with upregulation of lipogenic genes and downregulation of genes related to fatty acid oxidation. α-Cedrene stimulation resulted in a significant reduction in lipid contents of human hepatocytes and reprogramming of metabolic signatures, which are mediated by OR10J5, as demonstrated by receptor knockdown experiments using RNA interference. Taken together, our findings show a crucial role of OR10J5 in the regulation of lipid accumulation in human hepatocytes.

Integrated, High-Throughput, Multiomics Platform Enables Data-Driven Construction of Cellular Responses and Reveals Global Drug Mechanisms of Action

An understanding of how cells respond to perturbation is essential for biological applications; however, most approaches for profiling cellular response are limited in scope to pre-established targets. Global analysis of molecular mechanism will advance our understanding of the complex networks constituting cellular perturbation and lead to advancements in areas, such as infectious disease pathogenesis, developmental biology, pathophysiology, pharmacology, and toxicology. We have developed a high-throughput multiomics platform for comprehensive, de novo characterization of cellular mechanisms of action. Platform validation using cisplatin as a test compound demonstrates quantification of over 10 000 unique, significant molecular changes in less than 30 days. These data provide excellent coverage of known cisplatin-induced molecular changes and previously unrecognized insights into cisplatin resistance. This proof-of-principle study demonstrates the value of this platform as a resource to understand complex cellular responses in a high-throughput manner.

Functional characterization of somatic mutations in cancer using network-based inference of protein activity

Identifying the multiple dysregulated oncoproteins that contribute to tumorigenesis in a given patient is crucial for developing personalized treatment plans. However, accurate inference of aberrant protein activity in biological samples is still challenging as genetic alterations are only partially predictive and direct measurements of protein activity are generally not feasible. To address this problem we introduce and experimentally validate a new algorithm, VIPER (Virtual Inference of Protein-activity by Enriched Regulon analysis), for the accurate assessment of protein activity from gene expression data. We use VIPER to evaluate the functional relevance of genetic alterations in regulatory proteins across all TCGA samples. In addition to accurately inferring aberrant protein activity induced by established mutations, we also identify a significant fraction of tumors with aberrant activity of druggable oncoproteins—despite a lack of mutations, and vice-versa. In vitro assays confirmed that VIPER-inferred protein activity outperforms mutational analysis in predicting sensitivity to targeted inhibitors.

Effective gene expression data generation framework based on multi-model approach

OBJECTIVE

Overcome the lack of enough samples in gene expression data sets having thousands of genes but a small number of samples challenging the computational methods using them.

METHODS AND MATERIAL

This paper introduces a multi-model artificial gene expression data generation framework where different gene regulatory network (GRN) models contribute to the final set of samples based on the characteristics of their underlying paradigms. In the first stage, we build different GRN models, and sample data from each of them separately. Then, we pool the generated samples into a rich set of gene expression samples, and finally try to select the best of the generated samples based on a multi-objective selection method measuring the quality of the generated samples from three different aspects such as compatibility, diversity and coverage. We use four alternative GRN models, namely, ordinary differential equations, probabilistic Boolean networks, multi-objective genetic algorithm and hierarchical Markov model.

RESULTS

We conducted a comprehensive set of experiments based on both real-life biological and synthetic gene expression data sets. We show that our multi-objective sample selection mechanism effectively combines samples from different models having up to 95% compatibility, 10% diversity and 50% coverage. We show that the samples generated by our framework has up to 1.5x higher compatibility, 2x higher diversity and 2x higher coverage than the samples generated by the individual models that the multi-model framework uses. Moreover, the results show that the GRNs inferred from the samples generated by our framework can have 2.4x higher precision, 12x higher recall, and 5.4x higher f-measure values than the GRNs inferred from the original gene expression samples.

CONCLUSIONS

Therefore, we show that, we can significantly improve the quality of generated gene expression samples by integrating different computational models into one unified framework without dealing with complex internal details of each individual model. Moreover, the rich set of artificial gene expression samples is able to capture some biological relations that can even not be captured by the original gene expression data set.

Perturbation Detection Through Modeling of Gene Expression on a Latent Biological Pathway Network: A Bayesian hierarchical approach

Cellular response to a perturbation is the result of a dynamic system of biological variables linked in a complex network. A major challenge in drug and disease studies is identifying the key factors of a biological network that are essential in determining the cell's fate. Here our goal is the identification of perturbed pathways from high-throughput gene expression data. We develop a three-level hierarchical model, where (i) the first level captures the relationship between gene expression and biological pathways using confirmatory factor analysis, (ii) the second level models the behavior within an underlying network of pathways induced by an unknown perturbation using a conditional autoregressive model, and (iii) the third level is a spike-and-slab prior on the perturbations. We then identify perturbations through posterior-based variable selection. We illustrate our approach using gene transcription drug perturbation profiles from the DREAM7 drug sensitivity predication challenge data set. Our proposed method identified regulatory pathways that are known to play a causative role and that were not readily resolved using gene set enrichment analysis or exploratory factor models. Simulation results are presented assessing the performance of this model relative to a network-free variant and its robustness to inaccuracies in biological databases.

A method for analysis and design of metabolism using metabolomics data and kinetic models: Application on lipidomics using a novel kinetic model of sphingolipid metabolism

We present a model-based method, designated Inverse Metabolic Control Analysis (IMCA), which can be used in conjunction with classical Metabolic Control Analysis for the analysis and design of cellular metabolism. We demonstrate the capabilities of the method by first developing a comprehensively curated kinetic model of sphingolipid biosynthesis in the yeast Saccharomyces cerevisiae. Next we apply IMCA using the model and integrating lipidomics data. The combinatorial complexity of the synthesis of sphingolipid molecules, along with the operational complexity of the participating enzymes of the pathway, presents an excellent case study for testing the capabilities of the IMCA. The exceptional agreement of the predictions of the method with genome-wide data highlights the importance and value of a comprehensive and consistent engineering approach for the development of such methods and models. Based on the analysis, we identified the class of enzymes regulating the distribution of sphingolipids among species and hydroxylation states, with the D-phospholipase SPO14 being one of the most prominent. The method and the applications presented here can be used for a broader, model-based inverse metabolic engineering approach.

Inferring gene targets of drugs and chemical compounds from gene expression profiles

Motivation: Finding genes which are directly perturbed or targeted by drugs is of great interest and importance in drug discovery. Several network filtering methods have been created to predict the gene targets of drugs from gene expression data based on an ordinary differential equation model of the gene regulatory network (GRN). A critical step in these methods involves inferring the GRN from the expression data, which is a very challenging problem on its own. In addition, existing network filtering methods require computationally intensive parameter tuning or expression data from experiments with known genetic perturbations or both.

Results: We developed a method called DeltaNet for the identification of drug targets from gene expression data. Here, the gene target predictions were directly inferred from the data without a separate step of GRN inference. DeltaNet formulation led to solving an underdetermined linear regression problem, for which we employed least angle regression (DeltaNet-LAR) or LASSO regularization (DeltaNet-LASSO). The predictions using DeltaNet for expression data of Escherichia coli, yeast, fruit fly and human were significantly more accurate than those using network filtering methods, namely mode of action by network identification (MNI) and sparse simultaneous equation model (SSEM). Furthermore, DeltaNet using LAR did not require any parameter tuning and could provide computational speed-up over existing methods.

Conclusion: DeltaNet is a robust and numerically efficient tool for identifying gene perturbations from gene expression data. Importantly, the method requires little to no expert supervision, while providing accurate gene target predictions.

Availability and implementation: DeltaNet is available on http://www.cabsel.ethz.ch/tools/DeltaNet.

Contact:rudi.gunawan@chem.ethz.ch

Supplementary information:Supplementary data are available at Bioinformatics online.

Compound annotation with real time cellular activity profiles to improve drug discovery

INTRODUCTION

In the past decade, a range of innovative strategies have been developed to improve the productivity of pharmaceutical research and development. In particular, compound annotation, combined with informatics, has provided unprecedented opportunities for drug discovery.

AREAS COVERED

In this review, a literature search from 2000 to 2015 was conducted to provide an overview of the compound annotation approaches currently used in drug discovery. Based on this, a framework related to a compound annotation approach using real-time cellular activity profiles for probe, drug, and biology discovery is proposed.

EXPERT OPINION

Compound annotation with chemical structure, drug-like properties, bioactivities, genome-wide effects, clinical phenotypes, and textural abstracts has received significant attention in early drug discovery. However, these annotations are mostly associated with endpoint results. Advances in assay techniques have made it possible to obtain real-time cellular activity profiles of drug molecules under different phenotypes, so it is possible to generate compound annotation with real-time cellular activity profiles. Combining compound annotation with informatics, such as similarity analysis, presents a good opportunity to improve the rate of discovery of novel drugs and probes, and enhance our understanding of the underlying biology.

Synthetic biology for pharmaceutical drug discovery

Synthetic biology (SB) is an emerging discipline, which is slowly reorienting the field of drug discovery. For thousands of years, living organisms such as plants were the major source of human medicines. The difficulty in resynthesizing natural products, however, often turned pharmaceutical industries away from this rich source for human medicine. More recently, progress on transformation through genetic manipulation of biosynthetic units in microorganisms has opened the possibility of in-depth exploration of the large chemical space of natural products derivatives. Success of SB in drug synthesis culminated with the bioproduction of artemisinin by microorganisms, a tour de force in protein and metabolic engineering. Today, synthetic cells are not only used as biofactories but also used as cell-based screening platforms for both target-based and phenotypic-based approaches. Engineered genetic circuits in synthetic cells are also used to decipher disease mechanisms or drug mechanism of actions and to study cell-cell communication within bacteria consortia. This review presents latest developments of SB in the field of drug discovery, including some challenging issues such as drug resistance and drug toxicity.

Elucidating Compound Mechanism of Action by Network Perturbation Analysis

Genome-wide identification of the mechanism of action (MoA) of small-molecule compounds characterizing their targets, effectors, and activity modulators represents a highly relevant yet elusive goal, with critical implications for assessment of compound efficacy and toxicity. Current approaches are labor intensive and mostly limited to elucidating high-affinity binding target proteins. We introduce a regulatory network-based approach that elucidates genome-wide MoA proteins based on the assessment of the global dysregulation of their molecular interactions following compound perturbation. Analysis of cellular perturbation profiles identified established MoA proteins for 70% of the tested compounds and elucidated novel proteins that were experimentally validated. Finally, unknown-MoA compound analysis revealed altretamine, an anticancer drug, as an inhibitor of glutathione peroxidase 4 lipid repair activity, which was experimentally confirmed, thus revealing unexpected similarity to the activity of sulfasalazine. This suggests that regulatory network analysis can provide valuable mechanistic insight into the elucidation of small-molecule MoA and compound similarity.

Identification of Gene Networks with Time Delayed Regulation Based on Temporal Expression Profiles

There are fundamental limitations in inferring the functional interaction structure of a gene (regulatory) network only from sequence information such as binding motifs. To overcome such limitations, various approaches have been developed to infer the functional interaction structure from expression profiles. However, most of them have not been so successful due to the experimental limitations and computational complexity. Hence, there is a pressing need to develop a simple but effective methodology that can systematically identify the functional interaction structure of a gene network from time-series expression profiles. In particular, we need to take into account the different time delay effects in gene regulation since they are ubiquitously present. We have considered a new experiment that measures the overall expression changes after a perturbation on a specific gene. Based on this experiment, we have proposed a new inference method that can take account of the time delay induced while the perturbation affects its primary target genes. Specifically, we have developed an algebraic equation from which we can identify the subnetwork structure around the perturbed gene. We have also analyzed the influence of time delay on the inferred network structure. The proposed method is particularly useful for identification of a gene network with small variations in the time delay of gene regulation.

Control of cancer formation by intrinsic genetic noise and microenvironmental cues

Differentiation therapies that induce malignant cells to stop growing and revert to normal tissue-specific differentiated cell types are successful in the treatment of a few specific haematological tumours. However, this approach has not been widely applied to solid tumours because their developmental origins are less well understood. Recent advances suggest that understanding tumour cell plasticity and how intrinsic factors (such as genetic noise and microenvironmental signals, including physical cues from the extracellular matrix) govern cell state switches will help in the development of clinically relevant differentiation therapies for solid cancers.

Inferring Broad Regulatory Biology from Time Course Data: Have We Reached an Upper Bound under Constraints Typical of In Vivo Studies?

There is a growing appreciation for the network biology that regulates the coordinated expression of molecular and cellular markers however questions persist regarding the identifiability of these networks. Here we explore some of the issues relevant to recovering directed regulatory networks from time course data collected under experimental constraints typical of in vivo studies. NetSim simulations of sparsely connected biological networks were used to evaluate two simple feature selection techniques used in the construction of linear Ordinary Differential Equation (ODE) models, namely truncation of terms versus latent vector projection. Performance was compared with ODE-based Time Series Network Identification (TSNI) integral, and the information-theoretic Time-Delay ARACNE (TD-ARACNE). Projection-based techniques and TSNI integral outperformed truncation-based selection and TD-ARACNE on aggregate networks with edge densities of 10-30%, i.e. transcription factor, protein-protein cliques and immune signaling networks. All were more robust to noise than truncation-based feature selection. Performance was comparable on the in silico 10-node DREAM 3 network, a 5-node Yeast synthetic network designed for In vivo Reverse-engineering and Modeling Assessment (IRMA) and a 9-node human HeLa cell cycle network of similar size and edge density. Performance was more sensitive to the number of time courses than to sample frequency and extrapolated better to larger networks by grouping experiments. In all cases performance declined rapidly in larger networks with lower edge density. Limited recovery and high false positive rates obtained overall bring into question our ability to generate informative time course data rather than the design of any particular reverse engineering algorithm.

Conditional mutual inclusive information enables accurate quantification of associations in gene regulatory networks

Mutual information (MI), a quantity describing the nonlinear dependence between two random variables, has been widely used to construct gene regulatory networks (GRNs). Despite its good performance, MI cannot separate the direct regulations from indirect ones among genes. Although the conditional mutual information (CMI) is able to identify the direct regulations, it generally underestimates the regulation strength, i.e. it may result in false negatives when inferring gene regulations. In this work, to overcome the problems, we propose a novel concept, namely conditional mutual inclusive information (CMI2), to describe the regulations between genes. Furthermore, with CMI2, we develop a new approach, namely CMI2NI (CMI2-based network inference), for reverse-engineering GRNs. In CMI2NI, CMI2 is used to quantify the mutual information between two genes given a third one through calculating the Kullback-Leibler divergence between the postulated distributions of including and excluding the edge between the two genes. The benchmark results on the GRNs from DREAM challenge as well as the SOS DNA repair network in Escherichia coli demonstrate the superior performance of CMI2NI. Specifically, even for gene expression data with small sample size, CMI2NI can not only infer the correct topology of the regulation networks but also accurately quantify the regulation strength between genes. As a case study, CMI2NI was also used to reconstruct cancer-specific GRNs using gene expression data from The Cancer Genome Atlas (TCGA). CMI2NI is freely accessible at http://www.comp-sysbio.org/cmi2ni.

Data- and knowledge-based modeling of gene regulatory networks: an update

Gene regulatory network inference is a systems biology approach which predicts interactions between genes with the help of high-throughput data. In this review, we present current and updated network inference methods focusing on novel techniques for data acquisition, network inference assessment, network inference for interacting species and the integration of prior knowledge. After the advance of Next-Generation-Sequencing of cDNAs derived from RNA samples (RNA-Seq) we discuss in detail its application to network inference. Furthermore, we present progress for large-scale or even full-genomic network inference as well as for small-scale condensed network inference and review advances in the evaluation of network inference methods by crowdsourcing. Finally, we reflect the current availability of data and prior knowledge sources and give an outlook for the inference of gene regulatory networks that reflect interacting species, in particular pathogen-host interactions.

A systems approach to drug discovery in Alzheimer's disease

In the articles included in this volume, one feels a strong frustration among the writers with the slow course of therapeutics development for Alzheimer's disease and with the clinical failure of targeted therapeutic agents despite substantial progress in our understanding of the biology and biochemistry of the disease.

Distinguishing between driver and passenger mutations in individual cancer genomes by network enrichment analysis

Background

In somatic cancer genomes, delineating genuine driver mutations against a background of multiple passenger events is a challenging task. The difficulty of determining function from sequence data and the low frequency of mutations are increasingly hindering the search for novel, less common cancer drivers. The accumulation of extensive amounts of data on somatic point and copy number alterations necessitates the development of systematic methods for driver mutation analysis.

Results

We introduce a framework for detecting driver mutations via functional network analysis, which is applied to individual genomes and does not require pooling multiple samples. It probabilistically evaluates 1) functional network links between different mutations in the same genome and 2) links between individual mutations and known cancer pathways. In addition, it can employ correlations of mutation patterns in pairs of genes. The method was used to analyze genomic alterations in two TCGA datasets, one for glioblastoma multiforme and another for ovarian carcinoma, which were generated using different approaches to mutation profiling. The proportions of drivers among the reported de novo point mutations in these cancers were estimated to be 57.8% and 16.8%, respectively. The both sets also included extended chromosomal regions with synchronous duplications or losses of multiple genes. We identified putative copy number driver events within many such segments. Finally, we summarized seemingly disparate mutations and discovered a functional network of collagen modifications in the glioblastoma. In order to select the most efficient network for use with this method, we used a novel, ROC curve-based procedure for benchmarking different network versions by their ability to recover pathway membership.

Conclusions

The results of our network-based procedure were in good agreement with published gold standard sets of cancer genes and were shown to complement and expand frequency-based driver analyses. On the other hand, three sequence-based methods applied to the same data yielded poor agreement with each other and with our results. We review the difference in driver proportions discovered by different sequencing approaches and discuss the functional roles of novel driver mutations. The software used in this work and the global network of functional couplings are publicly available at http://research.scilifelab.se/andrej_alexeyenko/downloads.html.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2105-15-308) contains supplementary material, which is available to authorized users.

Silencing HoxA1 by intraductal injection of siRNA lipidoid nanoparticles prevents mammary tumor progression in mice

With advances in screening, the incidence of detection of premalignant breast lesions has increased in recent decades; however, treatment options remain limited to surveillance or surgical removal by lumpectomy or mastectomy. We hypothesized that disease progression could be blocked by RNA interference (RNAi) therapy and set out to develop a targeted therapeutic delivery strategy. Using computational gene network modeling, we identified HoxA1 as a putative driver of early mammary cancer progression in transgenic C3(1)-SV40TAg mice. Silencing this gene in cultured mouse or human mammary tumor spheroids resulted in increased acinar lumen formation, reduced tumor cell proliferation, and restoration of normal epithelial polarization. When the HoxA1 gene was silenced in vivo via intraductal delivery of nanoparticle-formulated small interfering RNA (siRNA) through the nipple of transgenic mice with early-stage disease, mammary epithelial cell proliferation rates were suppressed, loss of estrogen and progesterone receptor expression was prevented, and tumor incidence was reduced by 75%. This approach that leverages new advances in systems biology and nanotechnology offers a novel noninvasive strategy to block breast cancer progression through targeted silencing of critical genes directly within the mammary epithelium.

The inferred cardiogenic gene regulatory network in the mammalian heart

Cardiac development is a complex, multiscale process encompassing cell fate adoption, differentiation and morphogenesis. To elucidate pathways underlying this process, a recently developed algorithm to reverse engineer gene regulatory networks was applied to time-course microarray data obtained from the developing mouse heart. Approximately 200 genes of interest were input into the algorithm to generate putative network topologies that are capable of explaining the experimental data via model simulation. To cull specious network interactions, thousands of putative networks are merged and filtered to generate scale-free, hierarchical networks that are statistically significant and biologically relevant. The networks are validated with known gene interactions and used to predict regulatory pathways important for the developing mammalian heart. Area under the precision-recall curve and receiver operator characteristic curve are 9% and 58%, respectively. Of the top 10 ranked predicted interactions, 4 have already been validated. The algorithm is further tested using a network enriched with known interactions and another depleted of them. The inferred networks contained more interactions for the enriched network versus the depleted network. In all test cases, maximum performance of the algorithm was achieved when the purely data-driven method of network inference was combined with a data-independent, functional-based association method. Lastly, the network generated from the list of approximately 200 genes of interest was expanded using gene-profile uniqueness metrics to include approximately 900 additional known mouse genes and to form the most likely cardiogenic gene regulatory network. The resultant network supports known regulatory interactions and contains several novel cardiogenic regulatory interactions. The method outlined herein provides an informative approach to network inference and leads to clear testable hypotheses related to gene regulation.

Predicting dynamic signaling network response under unseen perturbations

MOTIVATION

Predicting trajectories of signaling networks under complex perturbations is one of the most valuable, but challenging, tasks in systems biology. Signaling networks are involved in most of the biological pathways, and modeling their dynamics has wide applications including drug design and treatment outcome prediction.

RESULTS

In this paper, we report a novel model for predicting the cell type-specific time course response of signaling proteins under unseen perturbations. This algorithm achieved the top performance in the 2013 8th Dialogue for Reverse Engineering Assessments and Methods (DREAM 8) subchallenge: time course prediction in breast cancer cell lines. We formulate the trajectory prediction problem into a standard regularization problem; the solution becomes solving this discrete ill-posed problem. This algorithm includes three steps: denoising, estimating regression coefficients and modeling trajectories under unseen perturbations. We further validated the accuracy of this method against simulation and experimental data. Furthermore, this method reduces computational time by magnitudes compared to state-of-the-art methods, allowing genome-wide modeling of signaling pathways and time course trajectories to be carried out in a practical time.

AVAILABILITY AND IMPLEMENTATION

Source code is available at http://guanlab.ccmb.med.umich.edu/DREAM/code.html and as supplementary file online.