Artificial neural networks modeling gene-environment interaction

ABSTRACT: BACKGROUND: Gene-environment interactions play an important role in the etiological pathway of complex diseases. An appropriate statistical method for handling a wide variety of complex situations involving interactions between variables is still lacking, especially when continuous variables are involved. The aim of this paper is to explore the ability of neural networks to model different structures of gene-environment interactions. A simulation study is set up to compare neural networks with standard logistic regression models. Eight different structures of gene-environment interactions are investigated. These structures are characterized by penetrance functions that are based on sigmoid functions or on combinations of linear and non-linear effects of a continuous environmental factor and a genetic factor with main effect or with a masking effect only. RESULTS: In our simulation study, neural networks are more successful in modeling gene-environment interactions than logistic regression models. This outperfomance is especially pronounced when modeling sigmoid penetrance functions, when distinguishing between linear and nonlinear components, and when modeling masking effects of the genetic factor. CONCLUSION: Our study shows that neural networks are a promising approach for analyzing gene-environment interactions. Especially, if no prior knowledge of the correct nature of the relationship between co-variables and response variable is present, neural networks provide a valuable alternative to regression methods that are limited to the analysis of linearly separable data.     

Prediction of protein--protein interaction sites in heterocomplexes with neural networks.

In this paper we address the problem of extracting features relevant for predicting protein--protein interaction sites from the three-dimensional structures of protein complexes. Our approach is based on information about evolutionary conservation and surface disposition. We implement a neural network based system, which uses a cross validation procedure and allows the correct detection of 73% of the residues involved in protein interactions in a selected database comprising 226 heterodimers. Our analysis confirms that the chemico-physical properties of interacting surfaces are difficult to distinguish from those of the whole protein surface. However neural networks trained with a reduced representation of the interacting patch and sequence profile are sufficient to generalize over the different features of the contact patches and to predict whether a residue in the protein surface is or is not in contact. By using a blind test, we report the prediction of the surface interacting sites of three structural components of the Dnak molecular chaperone system, and find close agreement with previously published experimental results. We propose that the predictor can significantly complement results from structural and functional proteomics.     

Pattern formation in the lateral line of zebrafish.

The lateral line of fish and amphibians is a sensory system that comprises a number of individual sense organs, the neuromasts, arranged in a defined pattern on the surface of the body. A conspicuous part of the system is a line of organs that extends along each flank (and which gave the system its name). At the end of zebrafish embryogenesis, this line comprises 7-8 neuromasts regularly spaced between the ear and the tip of the tail. The neuromasts are deposited by a migrating primordium that originates from the otic region. Here, we follow the development of this pattern and show that heterogeneities within the migrating primordium prefigure neuromast formation.     

Pattern formation by local self-activation and lateral inhibition

In 1972, we proposed a theory of biological pattern formation in which concentration maxima of pattern forming substances are generated through local self-enhancement in conjunction with long range inhibition. Since then, much evidence in various developmental systems has confirmed the importance of autocatalytic feedback loops combined with inhibitory interaction. Examples are found in the formation of embryonal organizing regions, in segmentation, in the polarization of individual cells, and in gene activation. By computer simulations, we have shown that the theory accounts for much of the regulatory phenomena observed, including signalling to regenerate removed parts. These self-regulatory features contribute to making development robust and error-tolerant. Furthermore, the resulting pattern is, to a large extent, independent of the details provided by initial conditions and inducing signals.     

Stability of dead zone bidirectional associative memory neural networks involving time delays

A mathematical model describing the dynamical interactions of bidirectional associative memory networks involving transmission delays is considered. The influence of a dead zone or a zone of noactivation on the global stability is investigated and various easily verifiable sets of sufficient conditions are established. The asymptotic nature of solutions when the given system of equations does not possess an equilibrium pattern is discussed.     

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.     

Pattern of functional extinctions in ecological networks with a variety of interaction types

There is a strong trend of declining populations in many species of both animals and plants. Dwindling numbers of species can eventually lead to their functional extinction. Functional, or ecological, extinction occurs when a species becomes too rare to fulfill its ecological, interactive role in the ecosystem, leading to true (numerical) extinction of other depending species. Recent theoretical work on food webs suggests that the frequency of functional extinction might be surprisingly high. However, little is known about the risk of functional species extinctions in networks with other types of interactions than trophic ones. Here, we explore the frequency of functional extinctions in model ecological networks having different proportions of antagonistic and mutualistic links. Furthermore, we investigate the topological relationship between functionally and numerically extinct species. We find that (1) the frequency of functional extinctions is higher in networks containing a mixture of antagonistic and mutualistic interactions than in networks with only one type of interaction, (2) increased mortality rate of species having both mutualistic and antagonistic links is more likely to lead to extinction of another species than to extinction of the species itself compared to species having only mutualistic or antagonistic links, and (3) trophic distance (shortest path) between functionally and numerically extinct species is, on average, longer than one, indicating the importance of indirect effects. These results generalize the findings of an earlier study on food webs, demonstrating the potential importance of functional extinction in a variety of ecological network types.     

Discovering protein complexes in protein interaction networks via exploring the weak ties effect

Background

Studying protein complexes is very important in biological processes since it helps reveal the structure-functionality relationships in biological networks and much attention has been paid to accurately predict protein complexes from the increasing amount of protein-protein interaction (PPI) data. Most of the available algorithms are based on the assumption that dense subgraphs correspond to complexes, failing to take into account the inherence organization within protein complex and the roles of edges. Thus, there is a critical need to investigate the possibility of discovering protein complexes using the topological information hidden in edges.

Results

To provide an investigation of the roles of edges in PPI networks, we show that the edges connecting less similar vertices in topology are more significant in maintaining the global connectivity, indicating the weak ties phenomenon in PPI networks. We further demonstrate that there is a negative relation between the weak tie strength and the topological similarity. By using the bridges, a reliable virtual network is constructed, in which each maximal clique corresponds to the core of a complex. By this notion, the detection of the protein complexes is transformed into a classic all-clique problem. A novel core-attachment based method is developed, which detects the cores and attachments, respectively. A comprehensive comparison among the existing algorithms and our algorithm has been made by comparing the predicted complexes against benchmark complexes.

Conclusions

We proved that the weak tie effect exists in the PPI network and demonstrated that the density is insufficient to characterize the topological structure of protein complexes. Furthermore, the experimental results on the yeast PPI network show that the proposed method outperforms the state-of-the-art algorithms. The analysis of detected modules by the present algorithm suggests that most of these modules have well biological significance in context of complexes, suggesting that the roles of edges are critical in discovering protein complexes.

     

Clustering with neural networks

Partitioning a set ofN patterns in ad-dimensional metric space intoK clusters — in a way that those in a given cluster are more similar to each other than the rest — is a problem of interest in many fields, such as, image analysis, taxonomy, astrophysics, etc. As there are approximatelyK ^N/K! possible ways of partitioning the patterns amongK clusters, finding the best solution is beyond exhaustive search whenN is large. We show that this problem, in spite of its exponential complexity, can be formulated as an optimization problem for which very good, but not necessarily optimal, solutions can be found by using a Hopfield model of neural networks. To obtain a very good solution, the network must start from many randomly selected initial states. The network is simulated on the MPP, a 128 × 128 SIMD array machine, where we use the massive parallelism not only in solving the differential equations that govern the evolution of the network, but also in starting the network from many initial states at once thus obtaining many solutions in one run. We achieve speedups of two to three orders of magnitude over serial implementations and the promise through Analog VLSI implementations of further speedups of three to six orders of magnitude.Supported by a National Research Council-NASA Research Associatship     

Pattern recognition in neural networks with competing dynamics: coexistence of fixed-point and cyclic attractors

We study the properties of the dynamical phase transition occurring in neural network models in which a competition between associative memory and sequential pattern recognition exists. This competition occurs through a weighted mixture of the symmetric and asymmetric parts of the synaptic matrix. Through a generating functional formalism, we determine the structure of the parameter space at non-zero temperature and near saturation (i.e., when the number of stored patterns scales with the size of the network), identifying the regions of high and weak pattern correlations, the spin-glass solutions, and the order-disorder transition between these regions. This analysis reveals that, when associative memory is dominant, smooth transitions appear between high correlated regions and spurious states. In contrast when sequential pattern recognition is stronger than associative memory, the transitions are always discontinuous. Additionally, when the symmetric and asymmetric parts of the synaptic matrix are defined in terms of the same set of patterns, there is a discontinuous transition between associative memory and sequential pattern recognition. In contrast, when the symmetric and asymmetric parts of the synaptic matrix are defined in terms of independent sets of patterns, the network is able to perform both associative memory and sequential pattern recognition for a wide range of parameter values.     

Fast adaptive formation of orthogonalizing filters and associative memory in recurrent networks of neuron-like elements

A demonstration is given that an orthogonalizing filter for patterns is formed adaptively and very rapidly in a network of neuron-like elements with internal feedback connections. It is here assumed that the feedback gain is variable, and proportional to the correlation matrix of the output pattern vectors. The time-dependent signal transfer properties of the complete system are described by a system matrix which satisfies a matrix Bernoulli differential equation; solutions of this equation are outlined. The asymptotic value of the system matrix is shown to correspond to the orthogonal projection operator on the space that is complementary to the space spanned by all of the earlier input pattern vectors. Such a system then acts as a filter, which optimally extracts the amount that is new in an input pattern with respect to all old patterns. It also has features that are directly attributable to a distributed associative memory that is optimally selective.     

Stability and attractivity in associative memory networks

We focus on stable and attractive states in a network having two-state neuron-like elements. We calculate the connection matrix which guarantees the stability and the strongest attractivity of p memorized patterns. We present an analytical evaluation of the patterns' attractivity. These results are illustrated by some computer simulations.     

Eye tracking using artificial neural networks for human computer interaction

This paper describes an ongoing project that has the aim to develop a low cost application to replace a computer mouse for people with physical impairment. The application is based on an eye tracking algorithm and assumes that the camera and the head position are fixed. Color tracking and template matching methods are used for pupil detection. Calibration is provided by neural networks as well as by parametric interpolation methods. Neural networks use back-propagation for learning and bipolar sigmoid function is chosen as the activation function. The user's eye is scanned with a simple web camera with backlight compensation which is attached to a head fixation device. Neural networks significantly outperform parametric interpolation techniques: 1) the calibration procedure is faster as they require less calibration marks and 2) cursor control is more precise. The system in its current stage of development is able to distinguish regions at least on the level of desktop icons. The main limitation of the proposed method is the lack of head-pose invariance and its relative sensitivity to illumination (especially to incidental pupil reflections).     

Serial order learning in associative formation

Serial order learning was investigated in rats by delivering food to two spatially distinct feeders. All rats received three food pellets in feeder A delivered 10 min into the session. Three additional pellets were then delivered in feeder B after 20s, in Group Short (S), or 150 s, in Group Long (L). The rats in Group S learned the A-B association better than the rats in Group L. Of more interest, however, was that the rats in Group S showed more anticipatory responding to feeder A, suggesting subjects had better learned that feeder A delivered food "first". Implications for classical conditioning and serial order learning are considered.     

GAP-43 mediates retinal axon interaction with lateral diencephalon cells during optic tract formation

GAP-43 is an abundant intracellular growth cone protein that can serve as a PKC substrate and regulate calmodulin availability. In mice with targeted disruption of the GAP-43 gene, retinal ganglion cell (RGC) axons fail to progress normally from the optic chiasm into the optic tracts. The underlying cause is unknown but, in principle, can result from either the disruption of guidance mechanisms that mediate axon exit from the midline chiasm region or defects in growth cone signaling required for entry into the lateral diencephalic wall to form the optic tracts. Results here show that, compared to wild-type RGC axons, GAP-43-deficient axons exhibit reduced growth in the presence of lateral diencephalon cell membranes. Reduced growth is not observed when GAP-43-deficient axons are cultured with optic chiasm, cortical, or dorsal midbrain cells. Lateral diencephalon cell conditioned medium inhibits growth of both wild-type and GAP-43-deficient axons to a similar extent and does not affect GAP-43-deficient axons more so. Removal or transplant replacement of the lateral diencephalon optic tract entry zone in GAP-43-deficient embryo preparations results in robust RGC axon exit from the chiasm. Together these data show that RGC axon exit from the midline region does not require GAP-43 function. Instead, GAP-43 appears to mediate RGC axon interaction with guidance cues in the lateral diencephalic wall, suggesting possible involvement of PKC and calmodulin signaling during optic tract formation.     

Influence of weak and covalent bonds on formation and hydrolysis of gelatin networks

The relative influence of physical and chemical bonds to overall gel properties are explored in gelatin gels. Physical, chemical, chemical-physical, and physical-chemical gels are obtained by cooling the protein solution and/or by transglutaminase reaction. Each type of network is characterized by rheology and polarimetry. It is shown that the overall properties as well as the dynamics inside the gels are dependent upon the order of formation and on the relative amount of triple helices and covalent bonds. Enzyme hydrolysis of covalent gels is slower than that of physical gels, as confirmed by the kinetics of helix release and degradation. A scheme is proposed to explain the results at both the physicochemical and the molecular levels.     

Concept formation in neural networks: Implications for evolution of cognitive functions

Parallel systems composed of many interconnected elements are both simple brain models and possible novel computer architectures. Potential advantages of such systems are massive parallelism with resulting speedup of computation as well as general ability to compute with noisy, corrupted, or missing data. Parallel, distributed, associative models have pronounced psychologies. Some ways of handling information are natural for them, and some things that we might want them to do are unnatural and quite difficult to do. A question of considerable interest is whether the models’ capabilities and limitations are features of human psychology. Such systems form categories based on the structure their inputs and display behavior that looks as if they form and use simple concepts. However, if noisy examples are learned, an initially stable concept structure may break up. One very simple function of names attached to categories — i.e. a rudimentary language — could be to stabilize a concept structure against fragmentation. In addition, if the statistical structure of the names reflects the statistical structure of the inputs, capacity and reliability of categorization and recognition is enhanced.     

Cardiovascular control with artificial neural networks

This paper describes the use of artificial neural networks to model cardiovascular autonomic control in a study of the hemodynamic changes associated with space flight. Cardiovascular system models were created including four parameters: heart rate, contractility, peripheral resistance, and venous tone. Artificial neural networks were then designed and trained. A technique known as backpropagation networking was used and the results of the application of this technique to heart rate control are presented and discussed.     

Antenna impedance matching with neural networks

Impedance matching between transmission lines and antennas is an important and fundamental concept in electromagnetic theory. One definition of antenna impedance is the resistance and reactance seen at the antenna terminals or the ratio of electric to magnetic fields at the input. The primary intent of this paper is real-time compensation for changes in the driving point impedance of an antenna due to frequency deviations. In general, the driving point impedance of an antenna or antenna array is computed by numerical methods such as the method of moments or similar techniques. Some configurations do lend themselves to analytical solutions, which will be the primary focus of this work. This paper employs a neural control system to match antenna feed lines to two common antennas during frequency sweeps. In practice, impedance matching is performed off-line with Smith charts or relatively complex formulas but they rarely perform optimally over a large bandwidth. There have been very few attempts to compensate for matching errors while the transmission system is in operation and most techniques have been targeted to a relatively small range of frequencies. The approach proposed here employs three small neural networks to perform real-time impedance matching over a broad range of frequencies during transmitter operation. Double stub tuners are being explored in this paper but the approach can certainly be applied to other methodologies. The ultimate purpose of this work is the development of an inexpensive microcontroller-based system.