The regulatory genome and the computer

The Genomic Code: the genome instantiates a generative model of the organism

How does the genome encode the form of the organism? What is the nature of this genomic code? Inspired by recent work in machine learning and neuroscience, we propose that the genome encodes a generative model of the organism. In this scheme, by analogy with variational autoencoders (VAEs), the genome comprises a connectionist network, embodying a compressed space of 'latent variables', with weights that get encoded by the learning algorithm of evolution and decoded through the processes of development. The generative model analogy accounts for the complex, distributed genetic architecture of most traits and the emergent robustness and evolvability of developmental processes, while also offering a conception that lends itself to formalization.

Form, function, mind: What doesn't compute (and what might)

The applicability of computational and dynamical systems models to organisms is scrutinized, using examples from developmental biology and cognition. Developmental morphogenesis is dependent on the inherent material properties of developing animal (metazoan) tissues, a non-computational modality, but cell differentiation, which utilizes chromatin-based revisable memory banks and program-like function-calling, via the developmental gene co-expression system unique to the metazoans, has a quasi-computational basis. Multi-attractor dynamical models are argued to be misapplied to global properties of development, and it is suggested that along with computationalism, classic forms of dynamicism are similarly unsuitable to accounting for cognitive phenomena. Proposals are made for treating brains and other nervous tissues as novel forms of excitable matter with inherent properties which enable the intensification of cell-based basal cognition capabilities present throughout the tree of life. Finally, some connections are drawn between the viewpoint described here and active inference models of cognition, such as the Free Energy Principle.

Cellular computation and cognition

Contemporary neural network models often overlook a central biological fact about neural processing: that single neurons are themselves complex, semi-autonomous computing systems. Both the information processing and information storage abilities of actual biological neurons vastly exceed the simple weighted sum of synaptic inputs computed by the "units" in standard neural network models. Neurons are eukaryotic cells that store information not only in synapses, but also in their dendritic structure and connectivity, as well as genetic "marking" in the epigenome of each individual cell. Each neuron computes a complex nonlinear function of its inputs, roughly equivalent in processing capacity to an entire 1990s-era neural network model. Furthermore, individual cells provide the biological interface between gene expression, ongoing neural processing, and stored long-term memory traces. Neurons in all organisms have these properties, which are thus relevant to all of neuroscience and cognitive biology. Single-cell computation may also play a particular role in explaining some unusual features of human cognition. The recognition of the centrality of cellular computation to "natural computation" in brains, and of the constraints it imposes upon brain evolution, thus has important implications for the evolution of cognition, and how we study it.

The machine-organism relation revisited

This article addresses some crucial assumptions that are rarely acknowledged when organisms and machines are compared. We begin by presenting a short historical reconstruction of the concept of "machine." We show that there has never been a unique and widely accepted definition of "machine" and that the extant definitions are based on specific technologies. Then we argue that, despite the concept's ambiguity, we can still defend a more robust, specific, and useful notion of machine analogy that accounts for successful strategies in connecting specific devices (or mechanisms) with particular living phenomena. For that purpose, we distinguish between what we call "generic identity" and proper "machine analogy." We suggest that "generic identity"-which, roughly stated, presumes that some sort of vague similarity might exist between organisms and machines-is a source of the confusion haunting many persistent disagreements and that, accordingly, it should be dismissed. Instead, we endorse a particular form of "machine analogy" where the relation between organic phenomena and mechanical devices is not generic but specific and grounded on the identification of shared "invariants." We propose that the machine analogy is a kind of analogy as proportion and we elucidate how this is used or might be used in scientific practices. We finally argue that while organisms are not machines in a generic sense, they might share many robust "invariants," which justify the scientists' use of machine analogies for grasping living phenomena.

Michael Waterman's Contributions to Computational Biology and Bioinformatics

On the occasion of Dr. Michael Waterman's 80th birthday, we review his major contributions to the field of computational biology and bioinformatics including the famous Smith-Waterman algorithm for sequence alignment, the probability and statistics theory related to sequence alignment, algorithms for sequence assembly, the Lander-Waterman model for genome physical mapping, combinatorics and predictions of ribonucleic acid structures, word counting statistics in molecular sequences, alignment-free sequence comparison, and algorithms for haplotype block partition and tagSNP selection related to the International HapMap Project. His books Introduction to Computational Biology: Maps, Sequences and Genomes for graduate students and Computational Genome Analysis: An Introduction geared toward undergraduate students played key roles in computational biology and bioinformatics education. We also highlight his efforts of building the computational biology and bioinformatics community as the founding editor of the Journal of Computational Biology and a founding member of the International Conference on Research in Computational Molecular Biology (RECOMB).

Are There Biomimetic Lessons from Genetic Regulatory Networks for Developing a Lunar Industrial Ecology?

We examine the prospect for employing a bio-inspired architecture for a lunar industrial ecology based on genetic regulatory networks. The lunar industrial ecology resembles a metabolic system in that it comprises multiple chemical processes interlinked through waste recycling. Initially, we examine lessons from factory organisation which have evolved into a bio-inspired concept, the reconfigurable holonic architecture. We then examine genetic regulatory networks and their application in the biological cell cycle. There are numerous subtleties that would be challenging to implement in a lunar industrial ecology but much of the essence of biological circuitry (as implemented in synthetic biology, for example) is captured by traditional electrical engineering design with emphasis on feedforward and feedback loops to implement robustness.

Genomic Intelligence as Über Bio-Cybersecurity: The Gödel Sentence in Immuno-Cognitive Systems

This paper gives formal foundations and evidence from gene science in the post Barbara McClintock era that the Gödel Sentence, far from being an esoteric construction in mathematical logic, is ubiquitous in genomic intelligence that evolved with multi-cellular life. Conditions uniquely found in the Adaptive Immune System (AIS) and Mirror Neuron System (MNS), termed the genomic immuno-cognitive system, coincide with three building blocks in computation theory of Gödel, Turing and Post (G-T-P). (i) Biotic elements have unique digital identifiers with gene codes executing 3D self-assembly for morphology and regulation of the organism using the recursive operation of Self-Ref (Self-Reference) with the other being a self-referential projection of self. (ii) A parallel offline simulation meta/mirror environment in 1–1 relation to online machine executions of self-codes gives G-T-P Self-Rep (Self-Representation). (iii) This permits a digital biotic entity to self-report that it is under attack by a biotic malware or non-self antigen in the format of the Gödel sentence, resulting in the “smarts” for contextual novelty production. The proposed unitary G-T-P recursive machinery in AIS and in MNS for social cognition yields a new explanation that the Interferon Gamma factor, known for friend-foe identification in AIS, is also integral to social behaviors. New G-T-P bio-informatics of AIS and novel anti-body production is given with interesting testable implications for COVID-19 pathology.

From axiomatic systems to the Dogmatic gene and beyond

The positivistic views that dominated the early debate on the foundations of mathematics, at the beginning of the 20th century, survived the "negative results" that have shown the limits of the axiomatic approach since the 1930s. Rigour, abstraction and symbolism have been confused with formalism, based on finite strings of signs, pre-given axioms, and potentially mechanisable rewriting rules. This contributed to major clarifications in the mathematical praxes but obscured the limits of formalisms due to the exclusion of the historical creation of sense proper to any science. We expand on this sometimes fruitful confusion with some case studies. We then hint to the historical creation of sense as a component of an epistemology of mathematics. We continue with an analogy with genocentric approaches in biology, as similar positivistic views resurfaced there fifty years later. Finite sequences of letters in the DNA would completely determine ontogenesis and phylogenesis, according to the Central Dogma of molecular biology. Limits and "negative evidence" have been disregarded while searching for the "gene for" everything. Alternative perspectives require a reconstruction of the sense of history as locus for the constitution of any object of biological knowledge. In particular, the historicity of biological evolution will be understood in terms of changing phase spaces and of the role of rare events in all phylogenetic trajectories. The analysis of the evolutionary production of variability, adaptivity and ecosystemic diversity is a key component of the project we hint to, as part of a renewed relation to the biological environment.

How haematopoiesis research became a fertile ground for regulatory network biology as pioneered by Eric Davidson

PURPOSE OF REVIEW

This historical perspective reviews how work of Eric H. Davidson was a catalyst and exemplar for explaining haematopoietic cell fate determination through gene regulation.

RECENT FINDINGS

Researchers studying blood and immune cells pioneered many of the early mechanistic investigations of mammalian gene regulatory processes. These efforts included the characterization of complex gene regulatory sequences exemplified by the globin and T-cell/B-cell receptor gene loci, as well as the identification of many key regulatory transcription factors through the fine mapping of chromosome translocation breakpoints in leukaemia patients. As the repertoire of known regulators expanded, assembly into gene regulatory network models became increasingly important, not only to account for the truism that regulatory genes do not function in isolation but also to devise new ways of extracting biologically meaningful insights from even more complex information. Here we explore how Eric H. Davidson's pioneering studies of gene regulatory network control in nonvertebrate model organisms have had an important and lasting impact on research into blood and immune cell development.

SUMMARY

The intellectual framework developed by Davidson continues to contribute to haematopoietic research, and his insistence on demonstrating logic and causality still challenges the frontier of research today.

Reawakening the Developmental Origins of Cancer Through Transposable Elements

Transposable elements (TEs) have an established role as important regulators of early human development, functioning as tissue-specific genes and regulatory elements. Functional TEs are highly active during early development, and interact with important developmental genes, some of which also function as oncogenes. Dedifferentiation is a hallmark of cancer, and is characterized by genetic and epigenetic changes that enable proliferation, self-renewal and a metabolism reminiscent of embryonic stem cells. There is also compelling evidence suggesting that the path to dedifferentiation in cancer can contribute to invasion and metastasis. TEs are frequently expressed in cancer, and recent work has identified a newly proposed mechanism involving extensive recruitment of TE-derived promoters to drive expression of oncogenes and subsequently promote oncogenesis—a process termed onco-exaptation. However, the mechanism by which this phenomenon occurs, and the extent to which it contributes to oncogenesis remains unknown. Initial hypotheses have proposed that onco-exaptation events are cancer-specific and arise randomly due to the dysregulated and hypomethylated state of cancer cells and abundance of TEs across the genome. However, we suspect that exaptation-like events may not just arise due to chance activation of novel regulatory relationships as proposed previously, but as a result of the reestablishment of early developmental regulatory relationships. Dedifferentiation in cancer is well-documented, along with expression of TEs. The known interactions between TEs and pluripotency factors such as NANOG and OCTt4 during early development, along with the expression of some placental-specific TE-derived transcripts in cancer support a possible link between TEs and dedifferentiation of tumor cells. Thus, we hypothesize that onco-exaptation events can be associated with the epigenetic reawakening of early developmental TEs to regulate expression of oncogenes and promote oncogenesis. We also suspect that activation of these early developmental regulatory TEs may promote dedifferentiation, although at this stage it is hard to predict whether TE activation is one of the initial drivers of dedifferentiation. We expect that developmental TE activation occurs as a result of the establishment of an epigenetic landscape in cancer that resembles that of early development and that developmental TE activation may also enable cancers to exploit early developmental pathways, repurposing them to promote malignancy.

Integrated mRNA, MicroRNA Transcriptome and Degradome Analyses Provide Insights into Stamen Development in Moso Bamboo

A flower is an essential organ for sexual reproduction in flowering plants, which has been extensively studied in model plants. In this study, we used transcriptomic, small RNA and degradome analyses to characterize key microRNAs (miRNAs) and their targets in floral organs of moso bamboo. In total, we identified 13,051 differentially expressed genes and 109 known miRNAs from 26 miRNA families. We aligned the miRNAs to known miRNA databases and revealed some conserved as well as novel miRNAs. Sixteen conserved miRNAs were specifically and highly expressed in stamens, including miRNA159 and miRNA166. In situ hybridization shows that miRNA159 plays a key role in the regulation of stamen development, and the expression levels of its targets PheMYB98 and PheMYB42 were low. Furthermore, Phe-MIRNA159 partially recovers phenotypes of mir159ab double mutant. Overexpression of Phe-MIR159 could cause failure in anther dehisce, and the mature pollens could not be dispersed and further reduce fertility in Arabidopsis. Semi-thin section result shows that anther endothelial layer of Phe-MIRNA159 overexpressing lines is lack of secondary thickening, resulting in limited force for anther opening. Phe-miR159 may regulate the expression of genes related to secondary thickening through negative regulation of AtMYB33, affecting the anther dehiscence. Taken together, this study provides insights regarding molecular networks underlying floral organs development of moso bamboo.

Profiling miRNA expression in photo-thermo-sensitive male genic sterility line (PTGMS) PA64S under high and low temperature

Peiai64S (PA64S) is a photo-thermo-sensitive genic male sterile line (PTGMS), with wide application in hybrid seed production in rice (Oryza sativa L.). Micro-RNAs are 21-24 nt, endogenously expressed small RNAs that have been characterized in various developmental stages of rice, but none have been studied with respect to the regulation of TGMS in rice. Here, we employed high-throughput sequencing to identify expression profiles of miRNAs in the anthers of PA64S at high (PA64S-H) and low temperature (PA64S-L). Two small RNA libraries from PA64S-H and PA64-L anthers were sequenced, and 263 known and 321 novel candidate miRNAs were identified. Based on the number of sequencing reads, a total of 133 known miRNAs were found to be differentially expressed between PA64S-H and PA64S-L. Target prediction showed that the target genes encode MYB and TCP transcription factors, and bHLH proteins. These target genes are related to pollen development and male sterility, suggesting that miRNA/targets may play roles in regulating TGMS in rice. Further, starch and sucrose metabolism pathways, sphingolipid metabolism, arginine and proline metabolism, and plant hormone signal transduction pathways were enriched by KEGG pathway annotation. These findings contribute to our understanding of the role of miRNAs during anther development and TGMS occurrence in rice.

Understanding Pattern Formation in Embryos: Experiment, Theory, and Simulation

Twenty-five years ago, Lewis Wolpert, the eminent developmental biologist, asked the question, "Do We Understand Development?" He concluded that such rapid progress had been made in the preceding two decades that "It is not unreasonable to think that enough will eventually be known to program a computer and simulate some aspects of development." This prediction has been fulfilled, at least partially, with data-driven simulations of several different developmental processes being developed in the intervening years. Nevertheless, the question remains of whether we "understand" development and if simulations are sufficient to provide an explanation of development. While in silico replications and models are undoubtedly an important tool in the investigation and dissection of developmental processes, which complement traditional experimental methods, these need to be supplemented by theory that identifies principles and provides coherent explanations. Here, I use the example of pattern formation in the vertebrate neural tube to illustrate this idea.

Mechanistic philosophies of development: Theodor Boveri and Eric H. Davidson

Theodor Boveri's (1862-1915) and Eric Davidson's (1937-2015) achievements represent thoroughly two quite distant time frames in the history of the mechanistic approaches to development, that Jane Maienschein (2014) has characterized respectively as the era of the "experimental embryo" and of the "computed embryo". Nonetheless, Davidson's special bond to Boveri is meant to emphasize the genealogical continuity of an embryological tradition of mechanistic philosophy that, differently from molecular biology, is committed to an explanation of the hereditary transmission of organization. Davidson's genealogical claim is reconsidered through a contextualized analysis of the function of machine-like models and of the role of experiment in the making of their respective mechanistic philosophies. This analysis may help to shed light on resilience and change in the understanding of a mechanistic approach to development.

Fitting structure to function in gene regulatory networks

Cascades of transcriptional regulation are the common source of the forward drive in all developmental systems. Increases in complexity and specificity of gene expression at successive stages are based on the collaboration of varied combinations of transcription factors already expressed in the cells to turn on new genes, and the logical relationships between the transcription factors acting and becoming newly expressed from stage to stage are best visualized as gene regulatory networks. However, gene regulatory networks used in different developmental contexts underlie processes that actually operate through different sets of rules, which affect the kinetics, synchronicity, and logical properties of individual network nodes. Contrasting early embryonic development in flies and sea urchins with adult mammalian hematopoietic development from stem cells, major differences are seen in transcription factor dosage dependence, the silencing or damping impacts of repression, and the impact of cellular regulatory history on the parts of the genome that are accessible to transcription factor action in a given cell type. These different features not only affect the kinds of models that can illuminate developmental mechanisms in the respective biological systems, but also reflect the evolutionary needs of these biological systems to optimize different aspects of development.

MicroRNA and Transcription Factor: Key Players in Plant Regulatory Network

Recent achievements in plant microRNA (miRNA), a large class of small and non-coding RNAs, are very exciting. A wide array of techniques involving forward genetic, molecular cloning, bioinformatic analysis, and the latest technology, deep sequencing have greatly advanced miRNA discovery. A tiny miRNA sequence has the ability to target single/multiple mRNA targets. Most of the miRNA targets are transcription factors (TFs) which have paramount importance in regulating the plant growth and development. Various families of TFs, which have regulated a range of regulatory networks, may assist plants to grow under normal and stress environmental conditions. This present review focuses on the regulatory relationships between miRNAs and different families of TFs like; NF-Y, MYB, AP2, TCP, WRKY, NAC, GRF, and SPL. For instance NF-Y play important role during drought tolerance and flower development, MYB are involved in signal transduction and biosynthesis of secondary metabolites, AP2 regulate the floral development and nodule formation, TCP direct leaf development and growth hormones signaling. WRKY have known roles in multiple stress tolerances, NAC regulate lateral root formation, GRF are involved in root growth, flower, and seed development, and SPL regulate plant transition from juvenile to adult. We also studied the relation between miRNAs and TFs by consolidating the research findings from different plant species which will help plant scientists in understanding the mechanism of action and interaction between these regulators in the plant growth and development under normal and stress environmental conditions.

Eric Davidson: Steps to a gene regulatory network for development

Eric Harris Davidson was a unique and creative intellectual force who grappled with the diversity of developmental processes used by animal embryos and wrestled them into an intelligible set of principles, then spent his life translating these process elements into molecularly definable terms through the architecture of gene regulatory networks. He took speculative risks in his theoretical writing but ran a highly organized, rigorous experimental program that yielded an unprecedentedly full characterization of a developing organism. His writings created logical order and a framework for mechanism from the complex phenomena at the heart of advanced multicellular organism development. This is a reminiscence of intellectual currents in his work as observed by the author through the last 30-35 years of Davidson's life.

Can functionality in evolving networks be explained reductively?

Philosophers of biology disagree about an adequate explication of the concept of function. Instead of perpetuating the debate on the level of in principle-arguments, this paper aims first at reconstructing functional talk in the biological research papers of Marom and Braun, which focus on two different kinds of evolving networks, and in discussing the ontological consequences which the authors draw from their results. Marom investigates evolving neural networks controlling Braitenberg vehicles. Braun observes the evolutionary rearrangement or "rewiring" of the genetic network of genetically modified yeast on a short time scale. In both cases, the parameters under investigation are defined in functional terms. However, both authors report striking differences in the structures that realize one and the same function, as well as striking differences in the function of identical structures. From this, they construct an argument against reductionism. The second aim of my paper is an inquiry into the epistemic legitimacy of this conclusion. This requires addressing critically several concepts on which Marom and Braun's argument is built.

Forced synchronization of autonomous dynamical Boolean networks

We present the design of an autonomous time-delay Boolean network realized with readily available electronic components. Through simulations and experiments that account for the detailed nonlinear response of each circuit element, we demonstrate that a network with five Boolean nodes displays complex behavior. Furthermore, we show that the dynamics of two identical networks display near-instantaneous synchronization to a periodic state when forced by a common periodic Boolean signal. A theoretical analysis of the network reveals the conditions under which complex behavior is expected in an individual network and the occurrence of synchronization in the forced networks. This research will enable future experiments on autonomous time-delay networks using readily available electronic components with dynamics on a slow enough time-scale so that inexpensive data collection systems can faithfully record the dynamics.

Comparative Study of Regulatory Circuits in Two Sea Urchin Species Reveals Tight Control of Timing and High Conservation of Expression Dynamics

Accurate temporal control of gene expression is essential for normal development and must be robust to natural genetic and environmental variation. Studying gene expression variation within and between related species can delineate the level of expression variability that development can tolerate. Here we exploit the comprehensive model of sea urchin gene regulatory networks and generate high-density expression profiles of key regulatory genes of the Mediterranean sea urchin, Paracentrotus lividus (Pl). The high resolution of our studies reveals highly reproducible gene initiation times that have lower variation than those of maximal mRNA levels between different individuals of the same species. This observation supports a threshold behavior of gene activation that is less sensitive to input concentrations. We then compare Mediterranean sea urchin gene expression profiles to those of its Pacific Ocean relative, Strongylocentrotus purpuratus (Sp). These species shared a common ancestor about 40 million years ago and show highly similar embryonic morphologies. Our comparative analyses of five regulatory circuits operating in different embryonic territories reveal a high conservation of the temporal order of gene activation but also some cases of divergence. A linear ratio of 1.3-fold between gene initiation times in Pl and Sp is partially explained by scaling of the developmental rates with temperature. Scaling the developmental rates according to the estimated Sp-Pl ratio and normalizing the expression levels reveals a striking conservation of relative dynamics of gene expression between the species. Overall, our findings demonstrate the ability of biological developmental systems to tightly control the timing of gene activation and relative dynamics and overcome expression noise induced by genetic variation and growth conditions.

Analytic approaches to stochastic gene expression in multicellular systems

Deterministic thermodynamic models of the complex systems, which control gene expression in metazoa, are helping researchers identify fundamental themes in the regulation of transcription. However, quantitative single cell studies are increasingly identifying regulatory mechanisms that control variability in expression. Such behaviors cannot be captured by deterministic models and are poorly suited to contemporary stochastic approaches that rely on continuum approximations, such as Langevin methods. Fortunately, theoretical advances in the modeling of transcription have assembled some general results that can be readily applied to systems being explored only through a deterministic approach. Here, I review some of the recent experimental evidence for the importance of genetically regulating stochastic effects during embryonic development and discuss key results from Markov theory that can be used to model this regulation. I then discuss several pairs of regulatory mechanisms recently investigated through a Markov approach. In each case, a deterministic treatment predicts no difference between the mechanisms, but the statistical treatment reveals the potential for substantially different distributions of transcriptional activity. In this light, features of gene regulation that seemed needlessly complex evolutionary baggage may be appreciated for their key contributions to reliability and precision of gene expression.

How to make a synthetic multicellular computer

Biological systems perform computations at multiple scales and they do so in a robust way. Engineering metaphors have often been used in order to provide a rationale for modeling cellular and molecular computing networks and as the basis for their synthetic design. However, a major constraint in this mapping between electronic and wet computational circuits is the wiring problem. Although wires are identical within electronic devices, they must be different when using synthetic biology designs. Moreover, in most cases the designed molecular systems cannot be reused for other functions. A new approximation allows us to simplify the problem by using synthetic cellular consortia where the output of the computation is distributed over multiple engineered cells. By evolving circuits in silico, we can obtain the minimal sets of Boolean units required to solve the given problem at the lowest cost using cellular consortia. Our analysis reveals that the basic set of logic units is typically non-standard. Among the most common units, the so called inverted IMPLIES (N-Implies) appears to be one of the most important elements along with the NOT and AND functions. Although NOR and NAND gates are widely used in electronics, evolved circuits based on combinations of these gates are rare, thus suggesting that the strategy of combining the same basic logic gates might be inappropriate in order to easily implement synthetic computational constructs. The implications for future synthetic designs, the general view of synthetic biology as a standard engineering domain, as well as potencial drawbacks are outlined.

Discrimination between thermodynamic models of cis-regulation using transcription factor occupancy data

Many studies have identified binding preferences for transcription factors (TFs), but few have yielded predictive models of how combinations of transcription factor binding sites generate specific levels of gene expression. Synthetic promoters have emerged as powerful tools for generating quantitative data to parameterize models of combinatorial cis-regulation. We sought to improve the accuracy of such models by quantifying the occupancy of TFs on synthetic promoters in vivo and incorporating these data into statistical thermodynamic models of cis-regulation. Using chromatin immunoprecipitation-seq, we measured the occupancy of Gcn4 and Cbf1 in synthetic promoter libraries composed of binding sites for Gcn4, Cbf1, Met31/Met32 and Nrg1. We measured the occupancy of these two TFs and the expression levels of all promoters in two growth conditions. Models parameterized using only expression data predicted expression but failed to identify several interactions between TFs. In contrast, models parameterized with occupancy and expression data predicted expression data, and also revealed Gcn4 self-cooperativity and a negative interaction between Gcn4 and Nrg1. Occupancy data also allowed us to distinguish between competing regulatory mechanisms for the factor Gcn4. Our framework for combining occupancy and expression data produces predictive models that better reflect the mechanisms underlying combinatorial cis-regulation of gene expression.

Population-specific regulation of Chmp2b by Lbx1 during onset of synaptogenesis in lateral association interneurons

Chmp2b is closely related to Vps2, a key component of the yeast protein complex that creates the intralumenal vesicles of multivesicular bodies. Dominant negative mutations in Chmp2b cause autophagosome accumulation and neurodegenerative disease. Loss of Chmp2b causes failure of dendritic spine maturation in cultured neurons. The homeobox gene Lbx1 plays an essential role in specifying postmitotic dorsal interneuron populations during late pattern formation in the neural tube. We have discovered that Chmp2b is one of the most highly regulated cell-autonomous targets of Lbx1 in the embryonic mouse neural tube. Chmp2b was expressed and depended on Lbx1 in only two of the five nascent, Lbx1-expressing, postmitotic, dorsal interneuron populations. It was also expressed in neural tube cell populations that lacked Lbx1 protein. The observed population-specific expression of Chmp2b indicated that only certain population-specific combinations of sequence specific transcription factors allow Chmp2b expression. The cell populations that expressed Chmp2b corresponded, in time and location, to neurons that make the first synapses of the spinal cord. Chmp2b protein was transported into neurites within the motor- and association-neuropils, where the first synapses are known to form between E11.5 and E12.5 in mouse neural tubes. Selective, developmentally-specified gene expression of Chmp2b may therefore be used to endow particular neuronal populations with the ability to mature dendritic spines. Such a mechanism could explain how mammalian embryos reproducibly establish the disynaptic cutaneous reflex only between particular cell populations.

Catching transcriptional regulation by thermostatistical modeling

Gene expression is frequently regulated by multiple transcription factors (TFs). Thermostatistical methods allow for a quantitative description of interactions between TFs, RNA polymerase and DNA, and their impact on the transcription rates. We illustrate three different scales of the thermostatistical approach: the microscale of TF molecules, the mesoscale of promoter energy levels and the macroscale of transcriptionally active and inactive cells in a cell population. We demonstrate versatility of combinatorial transcriptional activation by exemplifying logic functions, such as AND and OR gates. We discuss a metric for cell-to-cell transcriptional activation variability known as Fermi entropy. Suitability of thermostatistical modeling is illustrated by describing the experimental data on transcriptional induction of NFκB and the c-Fos protein.

The logic layout of the TOL network of Pseudomonas putida pWW0 plasmid stems from a metabolic amplifier motif (MAM) that optimizes biodegradation of m-xylene

Background

The genetic network of the TOL plasmid pWW0 of the soil bacterium Pseudomonas putida mt-2 for catabolism of m-xylene is an archetypal model for environmental biodegradation of aromatic pollutants. Although nearly every metabolic and transcriptional component of this regulatory system is known to an extraordinary molecular detail, the complexity of its architecture is still perplexing. To gain an insight into the inner layout of this network a logic model of the TOL system was implemented, simulated and experimentally validated. This analysis made sense of the specific regulatory topology out on the basis of an unprecedented network motif around which the entire genetic circuit for m-xylene catabolism gravitates.

Results

The most salient feature of the whole TOL regulatory network is the control exerted by two distinct but still intertwined regulators (XylR and XylS) on expression of two separated catabolic operons (upper and lower) for catabolism of m-xylene. Following model reduction, a minimal modular circuit composed by five basic variables appeared to suffice for fully describing the operation of the entire system. In silico simulation of the effect of various perturbations were compared with experimental data in which specific portions of the network were activated with selected inducers: m-xylene, o-xylene, 3-methylbenzylalcohol and 3-methylbenzoate. The results accredited the ability of the model to faithfully describe network dynamics. This analysis revealed that the entire regulatory structure of the TOL system enables the action an unprecedented metabolic amplifier motif (MAM). This motif synchronizes expression of the upper and lower portions of a very long metabolic system when cells face the head pathway substrate, m-xylene.

Conclusion

Logic modeling of the TOL circuit accounted for the intricate regulatory topology of this otherwise simple metabolic device. The found MAM appears to ensure a simultaneous expression of the upper and lower segments of the m-xylene catabolic route that would be difficult to bring about with a standard substrate-responsive single promoter. Furthermore, it is plausible that the MAM helps to avoid biochemical conflicts between competing plasmid-encoded and chromosomally-encoded pathways in this bacterium.

Algorithms in nature: the convergence of systems biology and computational thinking

Biologists rely on computational methods to analyze and integrate large data sets, while several computational methods were inspired by the high-level design principles of biological systems. This Perspectives discusses the recent convergence of these two ways of thinking.

Computer science and biology have enjoyed a long and fruitful relationship for decades. Biologists rely on computational methods to analyze and integrate large data sets, while several computational methods were inspired by the high-level design principles of biological systems. Recently, these two directions have been converging. In this review, we argue that thinking computationally about biological processes may lead to more accurate models, which in turn can be used to improve the design of algorithms. We discuss the similar mechanisms and requirements shared by computational and biological processes and then present several recent studies that apply this joint analysis strategy to problems related to coordination, network analysis, and tracking and vision. We also discuss additional biological processes that can be studied in a similar manner and link them to potential computational problems. With the rapid accumulation of data detailing the inner workings of biological systems, we expect this direction of coupling biological and computational studies to greatly expand in the future.

The conserved role and divergent regulation of foxa, a pan-eumetazoan developmental regulatory gene

Foxa is a forkhead transcription factor that is expressed in the endoderm lineage across metazoans. Orthologs of foxa are expressed in cells that intercalate, polarize, and form tight junctions in the digestive tracts of the mouse, the sea urchin, and the nematode and in the chordate notochord. The loss of foxa expression eliminates these morphogenetic processes. The remarkable similarity in foxa phenotypes in these diverse organisms raises the following questions: why is the developmental role of Foxa so highly conserved? Is foxa transcriptional regulation as conserved as its developmental role? Comparison of the regulation of foxa orthologs in sea urchin and in Caenorhabditis elegans shows that foxa transcriptional regulation has diverged significantly between these two organisms, particularly in the cells that contribute to the C. elegans pharynx formation. We suggest that the similarity of foxa phenotype is due to its role in an ancestral gene regulatory network that controlled intercalation followed by mesenchymal-to-epithelial transition. foxa transcriptional regulation had evolved to support the developmental program in each species so foxa would play its role controlling morphogenesis at the necessary embryonic address.

Implementing an OR-NOT (ORN) logic gate with components of the SOS regulatory network of Escherichia coli

Whether biological or electronic, man-engineered computation is based on logic circuits assembled with binary gates that are interconnected to perform Boolean operations. We report here the rewiring of the SOS system of Escherichia in a fashion that makes the output of both the recA and lexA promoters to faithfully follow the pattern of a binary composite OR-NOT gate (ORN) in which the inputs are DNA damage (e.g. nalidixic acid addition) and IPTG as an exogenous signal. Unlike other non-natural gates whose implementation requires changes in genes and promoters of the genome of the host cells, this ORN was brought about by the sole addition of wild-type bacteria with a plasmid encoding a module for LacI(q)-dependent expression of lexA. Specifically, we demonstrate that the interplay between native, chromosomally-encoded components of the SOS system and the extra parts engineered in such a plasmid made the desired performance to happen without any modification of the core DNA-damage response network. It is thus possible to artificially interface autonomous cell networks with a predetermined logic by means of Boolean gates built with regulatory elements already functioning in the recipient organism.

The logicome of environmental bacteria: merging catabolic and regulatory events with Boolean formalisms

The regulatory and metabolic networks that rule biodegradation of pollutants by environmental bacteria are wired to the rest of the cellular physiology through both transcriptional factors and intermediary signal molecules. In this review, we examine some formalisms for describing catalytic/regulatory circuits of this sort and advocate the adoption of Boolean logic for combining transcriptional and enzymatic occurrences in the same biological system. As an example, we show how known regulatory and metabolic actions that bring about biodegradation of m-xylene by Pseudomonas putida mt-2 can be represented as clusters of binary operations and then reconstructed as a digital network. Despite the many simplifications, Boolean tools still capture the gross behaviour of the system even in the absence of kinetic constants determined experimentally. On this basis, we argue that still with a limited volume of data binary formalisms allow us to penetrate the raison d'être of extant regulatory and metabolic architectures.

Functional conservation between rodents and chicken of regulatory sequences driving skeletal muscle gene expression in transgenic chickens

Background

Regulatory elements that control expression of specific genes during development have been shown in many cases to contain functionally-conserved modules that can be transferred between species and direct gene expression in a comparable developmental pattern. An example of such a module has been identified at the rat myosin light chain (MLC) 1/3 locus, which has been well characterised in transgenic mouse studies. This locus contains two promoters encoding two alternatively spliced isoforms of alkali myosin light chain. These promoters are differentially regulated during development through the activity of two enhancer elements. The MLC3 promoter alone has been shown to confer expression of a reporter gene in skeletal and cardiac muscle in transgenic mice and the addition of the downstream MLC enhancer increased expression levels in skeletal muscle. We asked whether this regulatory module, sufficient for striated muscle gene expression in the mouse, would drive expression in similar domains in the chicken.

Results

We have observed that a conserved downstream MLC enhancer is present in the chicken MLC locus. We found that the rat MLC1/3 regulatory elements were transcriptionally active in chick skeletal muscle primary cultures. We observed that a single copy lentiviral insert containing this regulatory cassette was able to drive expression of a lacZ reporter gene in the fast-fibres of skeletal muscle in chicken in three independent transgenic chicken lines in a pattern similar to the endogenous MLC locus. Reporter gene expression in cardiac muscle tissues was not observed for any of these lines.

Conclusions

From these results we conclude that skeletal expression from this regulatory module is conserved in a genomic context between rodents and chickens. This transgenic module will be useful in future investigations of muscle development in avian species.

Practical computational methods for regulatory genomics: a cisGRN-Lexicon and cisGRN-browser for gene regulatory networks

The CYRENE Project focuses on the study of cis-regulatory genomics and gene regulatory networks (GRN) and has three components: a cisGRN-Lexicon, a cisGRN-Browser, and the Virtual Sea Urchin software system. The project has been done in collaboration with Eric Davidson and is deeply inspired by his experimental work in genomic regulatory systems and gene regulatory networks. The current CYRENE cisGRN-Lexicon contains the regulatory architecture of 200 transcription factors encoding genes and 100 other regulatory genes in eight species: human, mouse, fruit fly, sea urchin, nematode, rat, chicken, and zebrafish, with higher priority on the first five species. The only regulatory genes included in the cisGRN-Lexicon (CYRENE genes) are those whose regulatory architecture is validated by what we call the Davidson Criterion: they contain functionally authenticated sites by site-specific mutagenesis, conducted in vivo, and followed by gene transfer and functional test. This is recognized as the most stringent experimental validation criterion to date for such a genomic regulatory architecture. The CYRENE cisGRN-Browser is a full genome browser tailored for cis-regulatory annotation and investigation. It began as a branch of the Celera Genome Browser (available as open source at http://sourceforge.net/projects/celeragb /) and has been transformed to a genome browser fully devoted to regulatory genomics. Its access paradigm for genomic data is zoom-to-the-DNA-base in real time. A more recent component of the CYRENE project is the Virtual Sea Urchin system (VSU), an interactive visualization tool that provides a four-dimensional (spatial and temporal) map of the gene regulatory networks of the sea urchin embryo.

Structural and regulatory evolution of cellular electrophysiological systems

Cellular electrophysiological systems, like developmental systems, appear to evolve primarily by means of regulatory evolution. It is suggested that electrophysiological systems share two key features with developmental systems that account for this dependence on regulatory evolution. For both systems, structural evolution has the potential to create significant problems of pleiotropy and both systems are predominantly computational in nature. It is concluded that the relative balance of physical and computational tasks that a biological system has to perform, combined with the probability that these tasks may have to change significantly during the course of evolution, will be major factors in determining the relative mix of regulatory and structural evolution that is observed for a given system. Physiological systems that directly interface with the environment will almost always perform some low-level physical task. In the majority of cases this will require evolution of protein function in order for the tasks themselves to evolve. For complex physiological systems a large fraction of their function will be devoted to high-level control functions that are predominantly computational in nature. In most cases regulatory evolution will be sufficient in order for these computational tasks to evolve.

Experimentally based sea urchin gene regulatory network and the causal explanation of developmental phenomenology

Gene regulatory networks for development underlie cell fate specification and differentiation. Network topology, logic and dynamics can be obtained by thorough experimental analysis. Our understanding of the gene regulatory network controlling endomesoderm specification in the sea urchin embryo has attained an advanced level such that it explains developmental phenomenology. Here we review how the network explains the mechanisms utilized in development to control the formation of dynamic expression patterns of transcription factors and signaling molecules. The network represents the genomic program controlling timely activation of specification and differentiation genes in the correct embryonic lineages. It can also be used to study evolution of body plans. We demonstrate how comparing the sea urchin gene regulatory network to that of the sea star and to that of later developmental stages in the sea urchin, reveals mechanisms underlying the origin of evolutionary novelty. The experimentally based gene regulatory network for endomesoderm specification in the sea urchin embryo provides unique insights into the system level properties of cell fate specification and its evolution.

On the basic computational structure of gene regulatory networks

Gene regulatory networks constitute the first layer of the cellular computation for cell adaptation and surveillance. In these webs, a set of causal relations is built up from thousands of interactions between transcription factors and their target genes. The large size of these webs and their entangled nature make it difficult to achieve a global view of their internal organisation. Here, this problem has been addressed through a comparative study of Escherichia coli, Bacillus subtilis and Saccharomyces cerevisiae gene regulatory networks. We extract the minimal core of causal relations, uncovering the hierarchical and modular organisation from a novel dynamical/causal perspective. Our results reveal a marked top-down hierarchy containing several small dynamical modules for E. coli and B. subtilis. Conversely, the yeast network displays a single but large dynamical module in the middle of a bow-tie structure. We found that these dynamical modules capture the relevant wiring among both common and organism-specific biological functions such as transcription initiation, metabolic control, signal transduction, response to stress, sporulation and cell cycle. Functional and topological results suggest that two fundamentally different forms of logic organisation may have evolved in bacteria and yeast.

Building developmental gene regulatory networks

Animal development is an elaborate process programmed by genomic regulatory instructions. Regulatory genes encode transcription factors and signal molecules, and their expression is under the control of cis-regulatory modules that define the logic of transcriptional responses to the inputs of other regulatory genes. The functional linkages among regulatory genes constitute the gene regulatory networks (GRNs) that govern cell specification and patterning in development. Constructing such networks requires identification of the regulatory genes involved and characterization of their temporal and spatial expression patterns. Interactions (activation/repression) among transcription factors or signals can be investigated by large-scale perturbation analysis, in which the function of each gene is specifically blocked. Resultant expression changes are then integrated to identify direct linkages, and to reveal the structure of the GRN. Predicted GRN linkages can be tested and verified by cis-regulatory analysis. The explanatory power of the GRN was shown in the lineage specification of sea urchin endomesoderm. Acquiring such networks is essential for a systematic and mechanistic understanding of the developmental process.

Modeling the dynamics of transcriptional gene regulatory networks for animal development

The dynamic process of cell fate specification is regulated by networks of regulatory genes. The architecture of the network defines the temporal order of specification events. To understand the dynamic control of the developmental process, the kinetics of mRNA and protein synthesis and the response of the cis-regulatory modules to transcription factor concentration must be considered. Here we review mathematical models for mRNA and protein synthesis kinetics which are based on experimental measurements of the rates of the relevant processes. The model comprises the response functions of cis-regulatory modules to their transcription factor inputs, by incorporating binding site occupancy and its dependence on biologically measurable quantities. We use this model to simulate gene expression, to distinguish between cis-regulatory execution of "AND" and "OR" logic functions, rationalize the oscillatory behavior of certain transcriptional auto-repressors and to show how linked subcircuits can be dealt with. Model simulations display the effects of mutation of binding sites, or perturbation of upstream gene expression. The model is a generally useful tool for understanding gene regulation and the dynamics of cell fate specification.

A combinatorial code for pattern formation in Drosophila oogenesis

Two-dimensional patterning of the follicular epithelium in Drosophila oogenesis is required for the formation of three-dimensional eggshell structures. Our analysis of a large number of published gene expression patterns in the follicle cells suggests that they follow a simple combinatorial code based on six spatial building blocks and the operations of union, difference, intersection, and addition. The building blocks are related to the distribution of inductive signals, provided by the highly conserved epidermal growth factor receptor and bone morphogenetic protein signaling pathways. We demonstrate the validity of the code by testing it against a set of patterns obtained in a large-scale transcriptional profiling experiment. Using the proposed code, we distinguish 36 distinct patterns for 81 genes expressed in the follicular epithelium and characterize their joint dynamics over four stages of oogenesis. The proposed combinatorial framework allows systematic analysis of the diversity and dynamics of two-dimensional transcriptional patterns and guides future studies of gene regulation.

The surprising complexity of the transcriptional regulation of the spdri gene reveals the existence of new linkages inside sea urchin's PMC and Oral Ectoderm Gene Regulatory Networks

During sea urchin embryogenesis the spdri gene participates in two separate Gene Regulatory Networks (GRNs): the Primary Mesenchyme Cells' (PMCs) and the Oral Ectoderm's one. In both cases, activation of the gene follows initial specification events [Amore, G., Yavrouian, R., Peterson, K., Ransick, A., McClay, D., Davidson, E., 2003. Spdeadringer, a sea urchin embryo gene required separately in skeletogenic and oral ectoderm gene regulatory networks. Dev. Biol. 261, 55-81.]. We identified a portion of genomic DNA ("4.7IL" -3456;+389) which is sufficient to replicate sdpri's expression pattern in experiments of transgenesis, using a GFP reporter. In our experiments, the activation kinetic of 4.7IL-GFP was similar to that of the endogenous gene and the reporter responded to known spdri's transcriptional regulators (Ets1, Alx1, Gsc and Dri). Here we present a dissection of this regulatory region and a description of the modules involved in spdri's transcriptional regulation. Both in the PMCs' and Oral Ectoderm's expression phases, activation of spdri is obtained through the integration of three kinds of inputs: positive and globally distributed ones; negative ones (that prevent ectopic expression); positive and tissue-specific ones. Our results allow to expand the map of the regulatory connections at the spdri node, both in the PMCs and in the Oral Ectoderm Gene Regulatory Networks (GRNs).

Dynamical patterning modules: physico-genetic determinants of morphological development and evolution

The shapes and forms of multicellular organisms arise by the generation of new cell states and types and changes in the numbers and rearrangements of the various kinds of cells. While morphogenesis and pattern formation in all animal species are widely recognized to be mediated by the gene products of an evolutionarily conserved 'developmental-genetic toolkit', the link between these molecular players and the physics underlying these processes has been generally ignored. This paper introduces the concept of 'dynamical patterning modules' (DPMs), units consisting of one or more products of the 'toolkit' genes that mobilize physical processes characteristic of chemically and mechanically excitable meso- to macroscopic systems such as cell aggregates: cohesion, viscoelasticity, diffusion, spatiotemporal heterogeneity based on lateral inhibition and multistable and oscillatory dynamics. We suggest that ancient toolkit gene products, most predating the emergence of multicellularity, assumed novel morphogenetic functions due to change in the scale and context inherent to multicellularity. We show that DPMs, acting individually and in concert with each other, constitute a 'pattern language' capable of generating all metazoan body plans and organ forms. The physical dimension of developmental causation implies that multicellular forms during the explosive radiation of animal body plans in the middle Cambrian, approximately 530 million years ago, could have explored an extensive morphospace without concomitant genotypic change or selection for adaptation. The morphologically plastic body plans and organ forms generated by DPMs, and their ontogenetic trajectories, would subsequently have been stabilized and consolidated by natural selection and genetic drift. This perspective also solves the apparent 'molecular homology-analogy paradox', whereby widely divergent modern animal types utilize the same molecular toolkit during development by proposing, in contrast to the Neo-Darwinian principle, that phenotypic disparity early in evolution occurred in advance of, rather than closely tracked, genotypic change.

Prediction of Cis-Regulatory Elements Controlling Genes Differentially Expressed by Retinal and Choroidal Vascular Endothelial Cells

Cultured endothelial cells of the human retina and choroid demonstrate distinct patterns of gene expression. We hypothesized that differential gene expression reflected differences in the interactions of transcription factors and respective cis-regulatory motifs(s) in these two emdothelial cell subpopulations, recognizing that motifs often exist as modules. We tested this hypothesis in silico by using TRANSFAC Professional and CisModule to identify cis-regulatory motifs and modules in genes that were differentially expressed by human retinal versus choroidal endothelial cells, as identified by analysis of a microarray data set. Motifs corresponding to eight transcription factors were significantly (p < 0.05) differentially abundant in genes that were relatively highly expressed in retinal (i.e., GCCR, HMGIY, HSF1, p53, VDR) or choroidal (i.e., E2F, YY1, ZF5) endothelial cells. Predicted cis-regulatory modules were quite different for these two groups of genes. Our findings raise the possibility of exploiting specific cis-regulatory motifs to target therapy at the ocular endothelial cells subtypes responsible for neovascular age-related macular degeneration or proliferative diabetic retinopathy.