Arpad Kelemen

Computational Intelligence in Bioinformatics

Bioinformatics involve the creation and advancement of algorithms using techniques including computational intelligence, applied mathematics and statistics, informatics, and biochemistry to solve biological problems usually on the molecular level. Major research efforts in the field include sequence analysis, gene finding, genome annotation, protein structure alignment analysis and prediction, prediction of gene expression, protein-protein docking/interactions, and the modeling of evolution. This book deals with the application of computational intelligence in bioinformatics. Addressing the various issues of bioinformatics using different computational intelligence approaches is the novelty of this edited volume.

Statistical advances and challenges for analyzing correlated high dimensional SNP data in genomic study for complex diseases

Recent advances of information technology in biomedical sciences and other applied areas have created numerous large diverse data sets with a high dimensional feature space, which provide us a tremendous amount of information and new opportunities for improving the quality of human life. Meanwhile, great challenges are also created driven by the continuous arrival of new data that requires researchers to convert these raw data into scientific knowledge in order to benefit from it. Association studies of complex diseases using SNP data have become more and more popular in biomedical research in recent years. In this paper, we present a review of recent statistical advances and challenges for analyzing correlated high dimensional SNP data in genomic association studies for complex diseases. The review includes both general feature reduction approaches for high dimensional correlated data and more specific approaches for SNPs data, which include unsupervised haplotype mapping, tag SNP selection, and supervised SNPs selection using statistical testing/scoring, statistical modeling and machine learning methods with an emphasis on how to identify interacting loci.

Brain atrophy and white matter hyperintensities in early Parkinson's disease†

The purpose of this research was to examine the extent of global brain atrophy and white matter hyperintensities (WMH) in early Parkinsons disease (PD) compared to normal controls (NC), to explore the relationship between the MRI variables and cognition in PD. In this multicenter study we included 155 PD patients (age 65.6 ± 9.1 years, disease duration 26.7 ± 19.9 months) and 101 age‐matched NC. On 3D‐T1‐WI, we calculated normalized brain volumes using SIENAX software. WMH volumes were assessed semiautomatically. In PD patients, correlation and regression analyses investigated the association between atrophy and WMH outcomes and global, attention‐executive, visuospatial, and memory cognitive functions. Regression analysis was controlled for age, education, depression score, motor severity, cerebrovascular risk, and sex. No significant MRI variable volume group differences were found. The models did not retain any of the imaging variables as significant predictors of cognitive impairment. There was no evidence of brain atrophy or higher WMH volume in PD compared to NC, and MRI volumetric measurements were not significant predictors of cognitive functions in PD patients. We conclude that global structural brain changes are not a major feature in patients with incident PD.

Associating phenotypes with molecular events: recent statistical advances and challenges underpinning microarray experiments

Progress in mapping the genome and developments in array technologies have provided large amounts of information for delineating the roles of genes involved in complex diseases and quantitative traits. Since complex phenotypes are determined by a network of interrelated biological traits typically involving multiple inter-correlated genetic and environmental factors that interact in a hierarchical fashion, microarrays hold tremendous latent information. The analysis of microarray data is, however, still a bottleneck. In this paper, we review the recent advances in statistical analyses for associating phenotypes with molecular events underpinning microarray experiments. Classical statistical procedures to analyze phenotypes in genetics are reviewed first, followed by descriptions of the statistical procedures for linking molecular events to measured gene expression phenotypes (microarray-based gene expression) and observed phenotypes such as diseases status. These statistical procedures include (1) prior analysis, such as data quality controls, and normalization analyses for minimizing the effects of experimental artifacts and random noise; (2) gene selections and differentiation procedures based on inferential statistics for the class comparisons; (3) dynamic temporal patterns analysis through exploratory statistics such as unsupervised clustering and supervised classification and predictions; (4) assessing the reliability of microarray studies using real-time PCR and the reproducibility issues from many studies and multiple platforms. In addition, the post analysis to associate the discovered patterns of gene expression to pathway and functional analysis for selected genes are also considered in order to increase our understanding of interconnected gene processes.

Differential and trajectory methods for time course gene expression data

MOTIVATION The issue of high dimensionality in microarray data has been, and remains, a hot topic in statistical and computational analysis. Efficient gene filtering and differentiation approaches can reduce the dimensions of data, help to remove redundant genes and noises, and highlight the most relevant genes that are major players in the development of certain diseases or the effect of drug treatment. The purpose of this study is to investigate the efficiency of parametric (including Bayesian and non-Bayesian, linear and non-linear), non-parametric and semi-parametric gene filtering methods through the application of time course microarray data from multiple sclerosis patients being treated with interferon-beta-1a. The analysis of variance with bootstrapping (parametric), class dispersion (semi-parametric) and Pareto (non-parametric) with permutation methods are presented and compared for filtering and finding differentially expressed genes. The Bayesian linear correlated model, the Bayesian non-linear model the and non-Bayesian mixed effects model with bootstrap were also developed to characterize the differential expression patterns. Furthermore, trajectory-clustering approaches were developed in order to investigate the dynamic patterns and inter-dependency of drug treatment effects on gene expression. RESULTS Results show that the presented methods performed significant differently but all were adequate in capturing a small number of the potentially relevant genes to the disease. The parametric method, such as the mixed model and two Bayesian approaches proved to be more conservative. This may because these methods are based on overall variation in expression across all time points. The semi-parametric (class dispersion) and non-parametric (Pareto) methods were appropriate in capturing variation in expression from time point to time point, thereby making them more suitable for investigating significant monotonic changes and trajectories of changes in gene expressions in time course microarray data. Also, the non-linear Bayesian model proved to be less conservative than linear Bayesian correlated growth models to filter out the redundant genes, although the linear model showed better fit than non-linear model (smaller DIC). We also report the trajectories of significant genes-since we have been able to isolate trajectories of genes whose regulations appear to be inter-dependent.

Temporal gene expression classification with regularised neural network

This paper proposes regularised neural networks for characterisation of the multiple heterogeneous temporal dynamic patterns of gene expressions. Regularisation is developed to deal with noisy, high dimensional time course data and overfitting problems. We test the proposed model with a popular gene expression data. The models performance is compared to other classification techniques, such as Nearest Neighbour, Support Vector Machine, and Self Organised Map. Results show that the proposed model can effectively capture the dynamic feature of gene expression temporal patterns despite the high noise levels, the highly correlated attributes, the overwhelming interactions, and other complex features typically present in microarray data.

Considerations for a successful clinical decision support system.

Clinical decision support systems have the potential to improve patient care in a multitude of ways. Clinical decision support systems can aid in the reduction of medical errors and reduction in adverse drug events, ensure comprehensive treatment of patient illnesses and conditions, encourage the adherence to guidelines, shorten patient length of stay, and decrease expenses over time. A clinical decision support system is one of the key components for reaching compliance for Meaningful Use. In this article, the advantages, potential drawbacks, and clinical decision support system adoption barriers are discussed, followed by an in-depth review of the characteristics that make a clinical decision support system successful. The legal and ethical issues that come with the implementation of a clinical decision support system within an organization and the future expectations of clinical decision support system are reviewed.

Bayesian neural network for microarray data

We propose Bayesian neural networks (BNN) with structural learning for exploring microarray data in gene expressions. The approach employs representative data and regularization to capture correlation among gene expressions and Bayesian techniques to extract gene expression information from noisy data. The performance was verified with stratified cross-validation and multiple iterated runs.

Optimizing intelligent agent's constraint satisfaction with neural networks

Finding suitable jobs for US Navy sailors from time to time is an important and ever-changing process. An Intelligent Distribution Agent and particularly its constraint satisfaction module take up the challenge to automate the process. The constraint satisfaction modules main task is to assign sailors to new jobs in order to maximize Navy and sailor happiness. We present various neural network techniques combined with several statistical criteria to optimize the modules performance and to make decisions in general. The data was taken from Navy databases and from surveys of Navy experts. Such indeterminate subjective component makes the optimization of the constraint satisfaction a very sophisticated task. Single-Layer Perceptron with logistic regression, Multilayer Perceptron with different structures and algorithms and Support Vector Machine with Adatron algorithm are presented for achieving best performance. Multilayer Perceptron neural network and Support Vector Machine with Adatron algorithm produced highly accurate classification and encouraging prediction.

Neuro-fuzzy classification for the job assignment problem

We propose to use an adaptive neuro-fuzzy inference system (ANFIS) in order to optimize decision making for the job assignment problem of the US Navy. Results need to be considered carefully because of the high noise level naturally present in the data coming from human decisions. We describe how the data was acquired and preprocessed. The design issues of the ANFIS are discussed.