CRBA: A Competitive Rate-Based Algorithm Based on Competitive Spiking Neural Networks.

In this paper we present a Competitive Rate-Based Algorithm (CRBA) that approximates operation of a Competitive Spiking Neural Network (CSNN). CRBA is based on modeling of the competition between neurons during a sample presentation, which can be reduced to ranking of the neurons based on a dot product operation and the use of a discrete Expectation Maximization algorithm; the latter is equivalent to the spike time-dependent plasticity rule. CRBA's performance is compared with that of CSNN on the MNIST and Fashion-MNIST datasets. The results show that CRBA performs on par with CSNN, while using three orders of magnitude less computational time. Importantly, we show that the weights and firing thresholds learned by CRBA can be used to initialize CSNN's parameters that results in its much more efficient operation.

SOTXTSTREAM: Density-based self-organizing clustering of text streams.

A streaming data clustering algorithm is presented building upon the density-based self-organizing stream clustering algorithm SOSTREAM. Many density-based clustering algorithms are limited by their inability to identify clusters with heterogeneous density. SOSTREAM addresses this limitation through the use of local (nearest neighbor-based) density determinations. Additionally, many stream clustering algorithms use a two-phase clustering approach. In the first phase, a micro-clustering solution is maintained online, while in the second phase, the micro-clustering solution is clustered offline to produce a macro solution. By performing self-organization techniques on micro-clusters in the online phase, SOSTREAM is able to maintain a macro clustering solution in a single phase. Leveraging concepts from SOSTREAM, a new density-based self-organizing text stream clustering algorithm, SOTXTSTREAM, is presented that addresses several shortcomings of SOSTREAM. Gains in clustering performance of this new algorithm are demonstrated on several real-world text stream datasets.

Self-Organizing Feature Maps Identify Proteins Critical to Learning in a Mouse Model of Down Syndrome.

Down syndrome (DS) is a chromosomal abnormality (trisomy of human chromosome 21) associated with intellectual disability and affecting approximately one in 1000 live births worldwide. The overexpression of genes encoded by the extra copy of a normal chromosome in DS is believed to be sufficient to perturb normal pathways and normal responses to stimulation, causing learning and memory deficits. In this work, we have designed a strategy based on the unsupervised clustering method, Self Organizing Maps (SOM), to identify biologically important differences in protein levels in mice exposed to context fear conditioning (CFC). We analyzed expression levels of 77 proteins obtained from normal genotype control mice and from their trisomic littermates (Ts65Dn) both with and without treatment with the drug memantine. Control mice learn successfully while the trisomic mice fail, unless they are first treated with the drug, which rescues their learning ability. The SOM approach identified reduced subsets of proteins predicted to make the most critical contributions to normal learning, to failed learning and rescued learning, and provides a visual representation of the data that allows the user to extract patterns that may underlie novel biological responses to the different kinds of learning and the response to memantine. Results suggest that the application of SOM to new experimental data sets of complex protein profiles can be used to identify common critical protein responses, which in turn may aid in identifying potentially more effective drug targets.

Simulating vertical and horizontal inhibition with short-term dynamics in a multi-column multi-layer model of neocortex.

The paper introduces a multi-layer multi-column model of the cortex that uses four different neuron types and short-term plasticity dynamics. It was designed with details of neuronal connectivity available in the literature and meets these conditions: (1) biologically accurate laminar and columnar flows of activity, (2) normal function of low-threshold spiking and fast spiking neurons, and (3) ability to generate different stages of epileptiform activity. With these characteristics the model allows for modeling lesioned or malformed cortex, i.e. examine properties of developmentally malformed cortex in which the balance between inhibitory neuron subtypes is disturbed.

Impact of HbA1c measurement on hospital readmission rates: analysis of 70,000 clinical database patient records.

Management of hyperglycemia in hospitalized patients has a significant bearing on outcome, in terms of both morbidity and mortality. However, there are few national assessments of diabetes care during hospitalization which could serve as a baseline for change. This analysis of a large clinical database (74 million unique encounters corresponding to 17 million unique patients) was undertaken to provide such an assessment and to find future directions which might lead to improvements in patient safety. Almost 70,000 inpatient diabetes encounters were identified with sufficient detail for analysis. Multivariable logistic regression was used to fit the relationship between the measurement of HbA1c and early readmission while controlling for covariates such as demographics, severity and type of the disease, and type of admission. Results show that the measurement of HbA1c was performed infrequently (18.4%) in the inpatient setting. The statistical model suggests that the relationship between the probability of readmission and the HbA1c measurement depends on the primary diagnosis. The data suggest further that the greater attention to diabetes reflected in HbA1c determination may improve patient outcomes and lower cost of inpatient care.

mi-DS: Multiple-Instance Learning Algorithm.

Multiple-instance learning (MIL) is a supervised learning technique that addresses the problem of classifying bags of instances instead of single instances. In this paper, we introduce a rule-based MIL algorithm, called mi-DS, and compare it with 21 existing MIL algorithms on 26 commonly used data sets. The results show that mi-DS performs on par with or better than several well-known algorithms and generates models characterized by balanced values of precision and recall. Importantly, the introduced method provides a framework that can be used for converting other rule-based algorithms into MIL algorithms.

Biological Restraint on the Izhikevich Neuron Model Essential for Seizure Modeling.

We propose a simple modification of the Izhikevich neuron model to restrict firing rates of neurons. We demonstrate how this modification affects overall network activity using a simple artificial network. Such restraint on the Izhikevich neuron model would be especially important in larger scale simulations or when frequency dependent short-term plasticity is one of the network components. Although maximum firing rates are most likely exceeded in simulations of seizure like activity or other conditions that promote excessive excitation, we show that restriction of neuronal firing frequencies has impact even on small networks with moderate levels of input.

Simulating lesions in multi-layer, multi-columnar model of neocortex.

The paper presents results of modeling global and focal loss of layers in a multi-columnar model of neocortex. Specifically, the spread of activity across columns in conditions of inhibitory blockade is compared. With very low inhibition activity spreads through all layers, however, deep layers are critical for spread of activity when inhibition is only moderately blocked.

Protein annotation from protein interaction networks and Gene Ontology.

We introduce a novel method for annotating protein function that combines Naïve Bayes and association rules, and takes advantage of the underlying topology in protein interaction networks and the structure of graphs in the Gene Ontology. We apply our method to proteins from the Human Protein Reference Database (HPRD) and show that, in comparison with other approaches, it predicts protein functions with significantly higher recall with no loss of precision. Specifically, it achieves 51% precision and 60% recall versus 45% and 26% for Majority and 24% and 61% for χ²-statistics, respectively.

Modeling of multisensory convergence with a network of spiking neurons: a reverse engineering approach.

Multisensory processing in the brain underlies a wide variety of perceptual phenomena, but little is known about the underlying mechanisms of how multisensory neurons are formed. This lack of knowledge is due to the difficulty for biological experiments to manipulate and test the parameters of multisensory convergence, the first and definitive step in the multisensory process. Therefore, by using a computational model of multisensory convergence, this study seeks to provide insight into the mechanisms of multisensory convergence. To reverse-engineer multisensory convergence, we used a biologically realistic neuron model and a biology-inspired plasticity rule, but did not make any a priori assumptions about multisensory properties of neurons in the network. The network consisted of two separate projection areas that converged upon neurons in a third area, and stimulation involved activation of one of the projection areas (or the other) or their combination. Experiments consisted of two parts: network training and multisensory simulation. Analyses were performed, first, to find multisensory properties in the simulated networks; second, to reveal properties of the network using graph theoretical approach; and third, to generate hypothesis related to the multisensory convergence. The results showed that the generation of multisensory neurons related to the topological properties of the network, in particular, the strengths of connections after training, was found to play an important role in forming and thus distinguishing multisensory neuron types.

Machine learning methods predict locomotor response to MK-801 in mouse models of down syndrome.

Down syndrome (DS), caused by trisomy of human chromosome 21 (HSA21), is a common genetic cause of cognitive impairment. This disorder results from the overexpression of HSA21 genes and the resulting perturbations in many molecular pathways and cellular processes. Knowledge-based identification of targets for pharmacotherapies will require defining the most critical protein abnormalities among these many perturbations. Here the authors show that using the Ts65Dn and Ts1Cje mouse models of DS, which are trisomic for 88 and 69 reference protein coding genes, respectively, a simple linear Naïve Bayes classifier successfully predicts behavioral outcome (level of locomotor activity) in response to treatment with the N-methyl-d-aspartate (NMDA) receptor antagonist MK-801. Input to the Naïve Bayes method were simple protein profiles generated from cortex and output was locomotor activity binned into three levels: low, medium, and high. When Feature Selection was used with the Naïve Bayes method, levels of three HSA21 and two non-HSA21 protein features were identified as making the most significant contributions to activity level. Using these five features, accuracies of up to 88% in prediction of locomotor activity were achieved. These predictions depend not only on genotype-specific differences but also on within-genotype individual variation in levels of molecular and behavioral parameters. With judicious choice of pathways and components, a similar approach may be useful in analysis of more complex behaviors, including those associated with learning and memory, and may facilitate identification of novel targets for pharmacotherapeutics.

Recognition of partially occluded and rotated images with a network of spiking neurons.

In this paper, we introduce a novel system for recognition of partially occluded and rotated images. The system is based on a hierarchical network of integrate-and-fire spiking neurons with random synaptic connections and a novel organization process. The network generates integrated output sequences that are used for image classification. The proposed network is shown to provide satisfactory predictive performance given that the number of the recognition neurons and synaptic connections are adjusted to the size of the input image. Comparison of synaptic plasticity activity rule (SAPR) and spike timing dependant plasticity rules, which are used to learn connections between the spiking neurons, indicates that the former gives better results and thus the SAPR rule is used. Test results show that the proposed network performs better than a recognition system based on support vector machines.

ClusFCM: an algorithm for predicting protein functions using homologies and protein interactions.

We introduce a new algorithm, called ClusFCM, which combines techniques of clustering and fuzzy cognitive maps (FCM) for prediction of protein functions. ClusFCM takes advantage of protein homologies and protein interaction network topology to improve low recall predictions associated with existing prediction methods. ClusFCM exploits the fact that proteins of known function tend to cluster together and deduce functions not only through their direct interaction with other proteins, but also from other proteins in the network. We use ClusFCM to annotate protein functions for Saccharomyces cerevisiae (yeast), Caenorhabditis elegans (worm), and Drosophila melanogaster (fly) using protein-protein interaction data from the General Repository for Interaction Datasets (GRID) database and functional labels from Gene Ontology (GO) terms. The algorithm's performance is compared with four state-of-the-art methods for function prediction--Majority, chi(2) statistics, Markov random field (MRF), and FunctionalFlow--using measures of Matthews correlation coefficient, harmonic mean, and area under the receiver operating characteristic (ROC) curves. The results indicate that ClusFCM predicts protein functions with high recall while not lowering precision. Supplementary information is available at http://www.egr.vcu.edu/cs/dmb/ClusFCM/.

A very simple spiking neuron model that allows for modeling of large, complex systems.

This letter introduces a biologically inspired very simple spiking neuron model. The model retains only crucial aspects of biological neurons: a network of time-delayed weighted connections to other neurons, a threshold-based generation of action potentials, action potential frequency proportional to stimulus intensity, and interneuron communication that occurs with time-varying potentials that last longer than the associated action potentials. The key difference between this model and existing spiking neuron models is its great simplicity: it is basically a collection of linear and discontinuous functions with no differential equations to solve. The model's ability to operate in a complex network was tested by using it as a basis of a network implementing a hypothetical echolocation system. The system consists of an emitter and two receivers. The outputs of the receivers are connected to a network of spiking neurons (using the proposed model) to form a detection grid that acts as a map of object locations in space. The network uses differences in the arrival times of the signals to determine the azimuthal angle of the source and time of flight to calculate the distance. The activation patterns observed indicate that for a network of spiking neurons, which uses only time delays to determine source locations, the spatial discrimination varies with the number and relative spacing of objects. These results are similar to those observed in animals that use echolocation.

A hidden Markov model for predicting protein interfaces.

Protein-protein interactions play a defining role in protein function. Identifying the sites of interaction in a protein is a critical problem for understanding its functional mechanisms, as well as for drug design. To predict sites within a protein chain that participate in protein complexes, we have developed a novel method based on the Hidden Markov Model, which combines several biological characteristics of the sequences neighboring a target residue: structural information, accessible surface area, and transition probability among amino acids. We have evaluated the method using 5-fold cross-validation on 139 unique proteins and demonstrated precision of 66% and recall of 61% in identifying interfaces. These results are better than those achieved by other methods used for identification of interfaces.

Prediction of protein secondary structure content for the twilight zone sequences.

Secondary protein structure carries information about local structural arrangements, which include three major conformations: alpha-helices, beta-strands, and coils. Significant majority of successful methods for prediction of the secondary structure is based on multiple sequence alignment. However, multiple alignment fails to provide accurate results when a sequence comes from the twilight zone, that is, it is characterized by low (<30%) homology. To this end, we propose a novel method for prediction of secondary structure content through comprehensive sequence representation, called PSSC-core. The method uses a multiple linear regression model and introduces a comprehensive feature-based sequence representation to predict amount of helices and strands for sequences from the twilight zone. The PSSC-core method was tested and compared with two other state-of-the-art prediction methods on a set of 2187 twilight zone sequences. The results indicate that our method provides better predictions for both helix and strand content. The PSSC-core is shown to provide statistically significantly better results when compared with the competing methods, reducing the prediction error by 5-7% for helix and 7-9% for strand content predictions. The proposed feature-based sequence representation uses a comprehensive set of physicochemical properties that are custom-designed for each of the helix and strand content predictions. It includes composition and composition moment vectors, frequency of tetra-peptides associated with helical and strand conformations, various property-based groups like exchange groups, chemical groups of the side chains and hydrophobic group, auto-correlations based on hydrophobicity, side-chain masses, hydropathy, and conformational patterns for beta-sheets. The PSSC-core method provides an alternative for predicting the secondary structure content that can be used to validate and constrain results of other structure prediction methods. At the same time, it also provides useful insight into design of successful protein sequence representations that can be used in developing new methods related to prediction of different aspects of the secondary protein structure.

Improved validation of peptide MS/MS assignments using spectral intensity prediction.

A major limitation in identifying peptides from complex mixtures by shotgun proteomics is the ability of search programs to accurately assign peptide sequences using mass spectrometric fragmentation spectra (MS/MS spectra). Manual analysis is used to assess borderline identifications; however, it is error-prone and time-consuming, and criteria for acceptance or rejection are not well defined. Here we report a Manual Analysis Emulator (MAE) program that evaluates results from search programs by implementing two commonly used criteria: 1) consistency of fragment ion intensities with predicted gas phase chemistry and 2) whether a high proportion of the ion intensity (proportion of ion current (PIC)) in the MS/MS spectra can be derived from the peptide sequence. To evaluate chemical plausibility, MAE utilizes similarity (Sim) scoring against theoretical spectra simulated by MassAnalyzer software (Zhang, Z. (2004) Prediction of low-energy collision-induced dissociation spectra of peptides. Anal. Chem. 76, 3908-3922) using known gas phase chemical mechanisms. The results show that Sim scores provide significantly greater discrimination between correct and incorrect search results than achieved by Sequest XCorr scoring or Mascot Mowse scoring, allowing reliable automated validation of borderline cases. To evaluate PIC, MAE simplifies the DTA text files summarizing the MS/MS spectra and applies heuristic rules to classify the fragment ions. MAE output also provides data mining functions, which are illustrated by using PIC to identify spectral chimeras, where two or more peptide ions were sequenced together, as well as cases where fragmentation chemistry is not well predicted.