Multistage interconnection networks (MINs) are popular in switching and communication applications and have been used in telecommunication and parallel computing systems for many years. Crosstalk a major problem introduced by an optical MIN, is caused by coupling two signals within a switching element. We focus on an efficient solution to avoiding crosstalk by routing traffic through an N×N optical network to avoid coupling two signals within each switching element using wavelength-division multiplexing (WDM) and a time-division approach. Under the constraint of avoiding crosstalk, the interest is on realizing a permutation that uses the minimum number of passes for routing. This routing problem is an NP-hard problem. Many heuristic algorithms are already designed by researchers to perform this routing such as a sequential algorithm, a degree-descending algorithm, etc. The genetic algorithm is used successfully to improve the performance over the heuristic algorithms. The drawback of the genetic algorithm is its long running times. We use the simulated annealing algorithm to improve the performance of solving the problem and optimizing the result. In addition, a wavelength lower bound estimate on the minimum number of passes required is calculated and compared to the results obtained using heuristic, genetic, and simulated annealing algorithms. Many cases are tested and the results are compared to the results of other algorithms to show the advantages of simulated annealing algorithm.
Protein secondary structure prediction is very important for drug design, protein engineering and immunological studies. This research uses fully connected multilayer perceptron (MLP) neural network with one, two and three hidden layers to predict protein secondary structure. Orthogonal matrix, BLOSUM62 matrix and hydrophobicity matrix are used for input profiles. To increase the input information for neural networks, the combined matrix from BLOSUM62 and orthogonal matrix and the combined matrix from BLOSUM62 and hydrophobicity matrix are also experimented. Binary classifiers indicate test accuracy of one hidden layer is better than that of two and three hidden layers. This may indicate that increasing complexity of architecture may not help neural network to recognize structural pattern of protein sequence more accurately. The results also show that the combined input profile of BLOSUM62 matrix and orthogonal matrix is the best one among five encoding schemes. While accuracy of the tertiary classifier reaches 63.20%, binary classifier for H/~H is 78.70%, which is comparable to other researchers’ results.
Over the decades, many studies have been done for the prediction of the protein structure. Since the protein secondary structure is closely related to the protein tertiary structure, many approaches begin with the prediction of secondary structure and apply the results to predict the tertiary structure. The recent trend of secondary structure prediction studies is mostly based on the neural network or the support vector machine (SVM). In this study, SVM is used as a machine learning tool for the prediction of secondary structure and several new encoding schemes, including orthogonal matrix, hydrophobicity matrix, BLOSUM62 substitution matrix and combined matrix of these, are developed and optimized to improve the prediction accuracy. Based on the best encoding scheme, each protein sequence is expressed as consecutive sliding windows and each amino acid inside a window is represented with 20 different matrix values. Once the optimal window length for six SVM binary classifiers is chosen to be 13 through many experiments, the new encoding scheme is tested based on this optimal window size with the 7-fold cross validation tests. The results show 2% increase in the accuracy of the binary classifiers when compared with the instances in which the classical orthogonal matrix is used. For the training and testing of the SVM binary classifiers, RS126 data sets is used since this is the common set adopted by the previous research groups. Finally, to combine the results of the six SVM binary classifiers, several existing tertiary classifiers are applied and the efficiency of each tertiary classifier is compared.
The prediction and modeling of protein structure is a central problem in bioinformatics. Neural networks have been used extensively to predict the secondary structure of proteins. While significant progress has been made by using multiple sequence data, the ability to predict secondary structure from a single sequence and a single prediction network has stagnated with an accuracy of about 75%. This implies that there is some limit to the accuracy of the prediction. In order to understand this behavior we asked the question of what happens as we change the target function for the prediction. Instead of predicting a derived quantity, such as whether a given chain is a helix, sheet or turn, we tested whether a more directly observed quantity such as the distance between a pair of α-carbon atoms could be predicted with reasonable accuracy. The α-carbon atom position is central to each residue in the protein and the distances between them in sequence define the backbone of protein. Knowledge of the distances between the α-carbon atoms is sufficient to determine the three dimensional structure of the protein. We have trained on distance data derived from the complete protein structure database (pdb) using a multi-layered perceptron feedforward neural network with back propagation. It shows that the root of mean square error is 0.4 Å with orthogonal coding of protein primary sequence. This is comparable to the experimental error in the structures used to form the database. The effects of exploring other encoding schemes, and different complexities of neural networks as well as related target functions such as distance thresholds will be presented.
Techniques based on neural networks can provide efficient solutions to a wide variety of problems in computer science. Routing in computer networks is to schedule messages and select communication links so that messages can be transferred efficiently between source and destination processors. Finding an optimal solution to many routing problems usually reqrueis exponential time and is impractical in reality. Hence, many heuristic algorithms have been designed to find sub-optimal solutions. In this research we use neural networks with a set of constraints to capture various collisions in multistage interconnection networks (MINs). Our simulation results have indicated that the Hopfield neural network can be used to routing to avoid link collisions in electronic MINs and crosstalks in optical MINs.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.