A formalisation of the neural assembly concept 1. Constraints on neural assembly size

Hebb proposed the concept of a neural assembly distributed across cortical tissue as a model for representation of information in the cerebral cortex. Later developments of the concept highlight the need for overlapping membership between independent assemblies, and the spread of activity throughout the assembly once it is activated above a critical level (ignition). Formalisation of the neural assembly concept, especially in relation to quantitative data from the real cortex, is at a very early stage. We consider two constraints on neural assembly size: (1) if a neural assembly is too small the fraction of its neurons that need to be active to ignite the whole assembly becomes unrealistically large; (2) if assemblies in a block of cortical tissue become too large then the block becomes ‘unsafe’, that is, unwanted spread from an active assembly to overlapping ones becomes inevitable. We consider variations in three parameters: neuronal firing threshold; connection density; and the total number of assemblies stored in the block of cortical tissue. Given biologically plausible values for these parameters we estimate maximum assembly size compatible with ignitability of individual assemblies, low probability of unwanted spread to overlapping assemblies, and safe operation of the block as a whole. Received: 7 March 1997 / Accepted in revised form: 1 July 1997     

GANN: Genetic algorithm neural networks for the detection of conserved combinations of features in DNA

Background  

The multitude of motif detection algorithms developed to date have largely focused on the detection of patterns in primary sequence. Since sequence-dependent DNA structure and flexibility may also play a role in protein-DNA interactions, the simultaneous exploration of sequence- and structure-based hypotheses about the composition of binding sites and the ordering of features in a regulatory region should be considered as well. The consideration of structural features requires the development of new detection tools that can deal with data types other than primary sequence.     

Axonal signals in the assembly of neural circuitry

Recent work in Drosophila and rodents has revealed that proteins transported along axons and delivered to pathway and target cell populations play important roles in the construction of neural circuitry. Interestingly, the parallels between these systems may extend to the identities of some of the molecules involved.     

Motifs in the assembly of food web networks

The assembly of local communities from regional pools is a multifaceted process that involves the confluence of interactions and environmental conditions at the local scale and biogeographic and evolutionary history at the regional scale. Understanding the relative influence of these factors on community structure has remained a challenge and mechanisms driving community assembly are often inferred from patterns of taxonomic, functional, and phylogenetic diversity. Moreover, community assembly is often viewed through the lens of competition and rarely includes trophic interactions or entire food webs. Here, we use motifs – subgraphs of nodes (e.g. species) and links (e.g. predation) whose abundance within a network deviates significantly as compared to a random network topology – to explore the assembly of food web networks found in the leaves of the northern pitcher plant Sarracenia purpurea. We compared counts of three‐node motifs across a hierarchy of scales to a suite of null models to determine if motifs are over‐, under‐, or randomly represented. We then assessed if the pattern of representation of a motif in a given network matched that of the network it was assembled from. We found that motif representation in over 70% of site networks matched the continental network they were assembled from and over 75% of local networks matched the site networks they were assembled from for the majority of null models. This suggests that the same processes are shaping networks across scales. To generalize our results and effectively use a motif perspective to study community assembly, a theoretical framework detailing potential mechanisms for all possible combinations of motif representation is necessary.     

Generalization mediates sensitivity to complex odor features in the honeybee

Animals use odors as signals for mate, kin, and food recognition, a strategy which appears ubiquitous and successful despite the high intrinsic variability of naturally-occurring odor quantities. Stimulus generalization, or the ability to decide that two objects, though readily distinguishable, are similar enough to afford the same consequence, could help animals adjust to variation in odor signals without losing sensitivity to key inter-stimulus differences. The present study was designed to investigate whether an animal's ability to generalize learned associations to novel odors can be influenced by the nature of the associated outcome. We use a classical conditioning paradigm for studying olfactory learning in honeybees to show that honeybees conditioned on either a fixed- or variable-proportion binary odor mixture generalize learned responses to novel proportions of the same mixture even when inter-odor differences are substantial. We also show that the resulting olfactory generalization gradients depend critically on both the nature of the stimulus-reward paradigm and the intrinsic variability of the conditioned stimulus. The reward dependency we observe must be cognitive rather than perceptual in nature, and we argue that outcome-dependent generalization is necessary for maintaining sensitivity to inter-odor differences in complex olfactory scenes.     

Markovian neural networks

The neural network that efficiently and nearly optimally solves difficult optimization problems is defined. The convergence proof for the Markovian neural network that asynchronously updates its neurons' states is also presented. The comparison of the performance of the Markovian neural network with various combinatorial optimization methods in two domains is described. The Markovian neural network is shown to be an efficient tool for solving optimization problems.     

Cephalopod neural networks

Cephalopods have arguably the largest and most complex nervous systems amongst the invertebrates; but despite the squid giant axon being one of the best studied nerve cells in neuroscience, and the availability of superb information on the morphology of some cephalopod brains, there is surprisingly little known about the operation of the neural networks that underlie the sophisticated range of behaviour these animals display. This review focuses on a few of the best studied neural networks: the giant fiber system, the chromatophore system, the statocyst system, the visual system and the learning and memory system, with a view to summarizing our current knowledge and stimulating new studies, particularly on the activities of identified central neurons, to provide a more complete understanding of networks within the cephalopod nervous system.     

Bayesian neural networks

A neural network that uses the basic Hebbian learning rule and the Bayesian combination function is defined. Analogously to Hopfield's neural network, the convergence for the Bayesian neural network that asynchronously updates its neurons' states is proved. The performance of the Bayesian neural network in four medical domains is compared with various classification methods. The Bayesian neural network uses more sophisticated combination function than Hopfield's neural network and uses more economically the available information. The naive Bayesian classifier typically outperforms the basic Bayesian neural network since iterations in network make too many mistakes. By restricting the number of iterations and increasing the number of fixed points the network performs better than the naive Bayesian classifier. The Bayesian neural network is designed to learn very quickly and incrementally.     

Feeding neural networks in the mollusc Aplysia

Aplysia feeding is striking in that it is executed with a great deal of plasticity. At least in part, this flexibility is a result of the organization of the feeding neural network. To illustrate this, we primarily discuss motor programs triggered via stimulation of the command-like cerebral-buccal interneuron 2 (CBI-2). CBI-2 is interesting in that it can generate motor programs that serve opposing functions, i.e., programs can be ingestive or egestive. When programs are egestive, radula-closing motor neurons are activated during the protraction phase of the motor program. When programs are ingestive, radula-closing motor neurons are activated during retraction. When motor programs change in nature, activity in the radula-closing circuitry is altered. Thus, CBI-2 stimulation stereotypically activates the protraction and retraction circuitry, with protraction being generated first, and retraction immediately thereafter. In contrast, radula-closing motor neurons can be activated during either protraction or retraction. Which will occur is determined by whether other cerebral and buccal neurons are recruited, e.g. radula-closing motor neurons tend to be activated during retraction if a second CBI, CBI-3, is recruited. Fundamentally different motor programs are, therefore, generated because CBI-2 activates some interneurons in a stereotypic manner and other interneurons in a variable manner.     

Oscillatory neural networks in the rabbit hippocampus

A model is described to account for damped oscillatory activity of two interacting neural populations, pyramidal cells and interneurons. This network in the hippocampus is treated as a lumped system with tine delays between elements. The physiological mechanism underlying the oscillatory activity appears to involve neural population interaction and cannot be described in terms of a network composed of but two neurons, a single pyramidal cell and a single interneuron. An unusual aspect of the model is the explicit incorporation of an ongoing background input to raise the mean level of activity of the pyramidal cell population. This model has evolved from a series of studies previously performed on cats. To test the model experiments were performed on rabbits. The data showing oscillatory activity following fornix stimulation in the rabbit indicate that the model can be applied not only to the cat but also to the rabbit. In additions, for commissural stimulation oscillatory potentials of neural populations and individual pyramidal cells were evoked as predicted by the model.     

Hypnosis modelling in neural networks

In the framework of the neural network theory effects similar to hypnotic displays are constructed. They are based on the associative paradigm involving non-linear interaction of excitatory and inhibitory channels with synaptic memory. The non-linearity of long-term memorizing processes may cause effects exhibited by blind spots, which are interpreted as the first stage of hypnosis. More complicated phenomena are discussed in terms of a two-layer network.     

Dynamic connections in neural networks

Massively parallel (neural-like) networks are receiving increasing attention as a mechanism for expressing information processing models. By exploiting powerful primitive units and stability-preserving construction rules, various workers have been able to construct and test quite complex models, particularly in vision research. But all of the detailed technical work was concerned with the structure and behavior offixed networks. The purpose of this paper is to extend the methodology to cover several aspects of change and memory.     

Theoretical approaches to holistic biological features: Pattern formation,neural networks and the brain-mind relation

The topic of this article is the relation between bottom-up and top-down, reductionist and “holistic” approaches to the solution of basic biological problems. While there is no doubt that the laws of physics apply to all events in space and time, including the domains of life, understanding biology depends not only on elucidating the role of the molecules involved, but, to an increasing extent, on systems theoretical approaches in diverse fields of the life sciences. Examples discussed in this article are the generation of spatial patterns in development by the interplay of autocatalysis and lateral inhibition; the evolution of integrating capabilities of the human brain, such as cognition-based empathy; and both neurobiological and epistemological aspects of scientific theories of consciousness and the mind.     

The nested assembly of individual-resource networks

1. Much of the current understanding of ecological systems is based on theory that does not explicitly take into account individual variation within natural populations. However, individuals may show substantial variation in resource use. This variation in turn may be translated into topological properties of networks that depict interactions among individuals and the food resources they consume (individual-resource networks). 2. Different models derived from optimal diet theory (ODT) predict highly distinct patterns of trophic interactions at the individual level that should translate into distinct network topologies. As a consequence, individual-resource networks can be useful tools in revealing the incidence of different patterns of resource use by individuals and suggesting their mechanistic basis. 3. In the present study, using data from several dietary studies, we assembled individual-resource networks of 10 vertebrate species, previously reported to show interindividual diet variation, and used a network-based approach to investigate their structure. 4. We found significant nestedness, but no modularity, in all empirical networks, indicating that (i) these populations are composed of both opportunistic and selective individuals and (ii) the diets of the latter are ordered as predictable subsets of the diets of the more opportunistic individuals. 5. Nested patterns are a common feature of species networks, and our results extend its generality to trophic interactions at the individual level. This pattern is consistent with a recently proposed ODT model, in which individuals show similar rank preferences but differ in their acceptance rate for alternative resources. Our findings therefore suggest a common mechanism underlying interindividual variation in resource use in disparate taxa.     

Ensembling multiple raw coevolutionary features with deep residual neural networks for contact-map prediction in CASP13

We report the results of residue-residue contact prediction of a new pipeline built purely on the learning of coevolutionary features in the CASP13 experiment. For a query sequence, the pipeline starts with the collection of multiple sequence alignments (MSAs) from multiple genome and metagenome sequence databases using two complementary Hidden Markov Model (HMM)-based searching tools. Three profile matrices, built on covariance, precision, and pseudolikelihood maximization respectively, are then created from the MSAs, which are used as the input features of a deep residual convolutional neural network architecture for contact-map training and prediction. Two ensembling strategies have been proposed to integrate the matrix features through end-to-end training and stacking, resulting in two complementary programs called TripletRes and ResTriplet, respectively. For the 31 free-modeling domains that do not have homologous templates in the PDB, TripletRes and ResTriplet generated comparable results with an average accuracy of 0.640 and 0.646, respectively, for the top L/5 long-range predictions, where 71% and 74% of the cases have an accuracy above 0.5. Detailed data analyses showed that the strength of the pipeline is due to the sensitive MSA construction and the advanced strategies for coevolutionary feature ensembling. Domain splitting was also found to help enhance the contact prediction performance. Nevertheless, contact models for tail regions, which often involve a high number of alignment gaps, and for targets with few homologous sequences are still suboptimal. Development of new approaches where the model is specifically trained on these regions and targets might help address these problems.