Răzvan Andonie

Fuzzy ARTMAP Prediction of Biological Activities for Potential HIV-1 Protease Inhibitors Using a Small Molecular Data Set

Obtaining satisfactory results with neural networks depends on the availability of large data samples. The use of small training sets generally reduces performance. Most classical Quantitative Structure-Activity Relationship (QSAR) studies for a specific enzyme system have been performed on small data sets. We focus on the neuro-fuzzy prediction of biological activities of HIV-1 protease inhibitory compounds when inferring from small training sets. We propose two computational intelligence prediction techniques which are suitable for small training sets, at the expense of some computational overhead. Both techniques are based on the FAMR model. The FAMR is a Fuzzy ARTMAP (FAM) incremental learning system used for classification and probability estimation. During the learning phase, each sample pair is assigned a relevance factor proportional to the importance of that pair. The two proposed algorithms in this paper are: 1) The GA-FAMR algorithm, which is new, consists of two stages: a) During the first stage, we use a genetic algorithm (GA) to optimize the relevances assigned to the training data. This improves the generalization capability of the FAMR. b) In the second stage, we use the optimized relevances to train the FAMR. 2) The Ordered FAMR is derived from a known algorithm. Instead of optimizing relevances, it optimizes the order of data presentation using the algorithm of Dagher et al. In our experiments, we compare these two algorithms with an algorithm not based on the FAM, the FS-GA-FNN introduced in . We conclude that when inferring from small training sets, both techniques are efficient, in terms of generalization capability and execution time. The computational overhead introduced is compensated by better accuracy. Finally, the proposed techniques are used to predict the biological activities of newly designed potential HIV-1 protease inhibitors.

Neuro-fuzzy Prediction of Biological Activity and Rule Extraction for HIV-1 Protease Inhibitors

A fuzzy neural network (FNN) and multiple linear regression (MLR) were used to predict biological activities of 26 newly designed HIV-1 protease potential inhibitory compounds. Molecular descriptors of 151 known inhibitors were used to train and test the FNN and to develop MLR models. The predictive ability of these two models was investigated and compared. We found the predictive ability of the FNN to be generally superior to that of MLR. The fuzzy IF/THEN rules were extracted from the trained network. These rules map chemical structure descriptors to predicted inhibitory values. The obtained rules can be used to analyze the influence of descriptors. Our results indicate that FNN and fuzzy IF/THEN rules are powerful modeling tools for QSAR studies.

Implementation Issues of an Incremental and Decremental SVM

Incremental and decremental processes of training a support vector machine (SVM) resumes to the migration of vectors in and out of the support set along with modifying the associated thresholds. This paper gives an overview of all the boundary conditions implied by vector migration through the incremental / decremental process. The analysis will show that the same procedures, with very slight variations, can be used for both the incremental and decremental learning. The case of vectors with duplicate contribution is also considered. Migration of vectors among sets on decreasing the regularization parameter is given particularly attention. Experimental data show the possibility of modifying this parameter on a large scale, varying it from complete training (overfitting) to a calibrated value.

An Integrated Soft Computing Approach for Predicting Biological Activity of Potential HIV-1 Protease Inhibitors

Using a neural network-fuzzy logic-genetic algorithm approach we generate an optimal predictor for biological activities of HIV-1 protease potential inhibitory compounds. We use genetic algorithms (GAs) in the two optimization stages. In the first stage, we generate an optimal subset of features. In the second stage, we optimize the architecture of the fuzzy neural network. The optimized network is trained and used for the prediction of biological activities of newly designed chemical compounds. Finally, we extract fuzzy IF/THEN rules. These rules map physico-chemical structure descriptors to predicted inhibitory values. The optimal subset of features, combined with the generated rules, can be used to analyze the influence of descriptors.

An Implementation of Evolutionary Computation Operators in OpenCL

We discuss the parallel implementation of Genetic Algorithms and Evolution Strategy on General-Purpose Graphical Units, using the OpenCL framework. Multiple evolutionary operators are tested (tournament, roulette wheel selection, uniform and Gaussian mutation, crossover, recombination), as well as different approaches for parallelism, for small and large problem sizes. We use the Island Model of Parallel GA, with random migration. Performance is measured using two graphic cards: NVidia GeForce GTX 560Ti and AMD Radeon 6950. Tests are performed in a distributed grid, using the Java Parallel Processing Framework.

Informational energy kernel for LVQ

We describe a kernel method which uses the maximization of Onicescus informational energy as a criteria for computing the relevances of input features. This adaptive relevance determination is used in combination with the neural-gas and the generalized relevance LVQ algorithms. Our quadratic optimization function, as an L2 type method, leads to linear gradient and thus easier computation. We obtain an approximation formula similar to the mutual information based method, but in a more simple way.

Clustering and visualization of geodetic array data streams using self-organizing maps

The Pacific Northwest Geodesic Array at Central Washington University collects telemetered streaming data from 450 GPS stations. These real-time data are used to monitor and mitigate natural hazards arising from earthquakes, volcanic eruptions, landslides, and coastal sea-level hazards in the Pacific Northwest. Recent improvements in both accuracy of positioning measurements and latency of terrestrial data communication have led to the ability to collect data with higher sampling rates. For seismic monitoring applications, this means 1350 separate position streams from stations located across 1200 km along the West Coast of North America must be able to be both visually observed and automatically analyzed at a sampling rate of up to 1 Hz. Our goal is to efficiently extract and visualize useful information from these data streams. We propose a method to visualize the geodetic data by clustering the signal types with a Self-Organizing Map (SOM). The similarity measure in the SOM is determined by the similarity of signals received from GPS stations. Signals are transformed to symbol strings, and the distance measure in the SOM is defined by an edit distance. The symbol strings represent data streams and the SOM is dynamic. We overlap the resulted dynamic SOM on the Google Maps representation.

How to infer the informational energy from small datasets

Motivated by the problems in machine learning, we introduce a novel non-parametric estimator of Onicescus informational energy. Our method is based on the k-th nearest neighbor distances between the n sample points, where k is a fixed positive integer. For some standard distributions, we investigate the performance of the estimator for small datasets.

Energy Supervised Relevance Neural Gas for Feature Ranking

In pattern classification, input pattern features usually contribute differently, in accordance to their relevances for a specific classification task. In a previous paper, we have introduced the Energy Supervised Relevance Neural Gas classifier, a kernel method which uses the maximization of Onicescu’s informational energy for computing the relevances of input features. Relevances were used to improve classification accuracy. In our present work, we focus on the feature ranking capability of this approach. We compare our algorithm to standard feature ranking methods.

Using kolmogorov inspired gates for low power nanoelectronics

Based on explicit numerical constructions for Kolmogorovs superpositions (KS) linear size circuits are possible. Because classical Boolean as well as threshold logic implementations require exponential size in the worst case, it follows that size-optimal solutions for arbitrary Boolean functions (BFs) should rely (at least partly) on KS. In this paper, we will present previous theoretical results while examining the particular case of 3-input BFs in detail. This shows that there is still room for improvement on the synthesis of BFs. Such size reductions (which can be achieved systematically) could help alleviate the challenging power consumption problem, and advocate for the design of Kolmogorov-inspired gates, as well as for the development of the theory, the algorithms, and the CAD tools that would allow taking advantage of such optimal combinations of different logic styles.