Alioune Ngom

Genetic algorithm based scheduler for computational grids

In the context of highly scalable distributed resource management architectures for grid computing, we present a genetic algorithm based scheduler. A scheduler must use the available resources efficiently, while satisfying competing and mutually conflicting goals. The grid workload may consist of multiple jobs, with quality-of-service constraints. A directed acyclic graph (DAG) represents each job, taking into account arbitrary precedence constraints and arbitrary processing time. The scheduler has been designed to be compatible with other tools being developed by our grid research group. We present the design, implementation and test results for such a scheduler in which we minimize make-span, idle time of the available computational resources, turn-around time and the specified deadlines provided by users. The architecture is hierarchical and the scheduler is usable at either the lowest or the higher tiers. It can also be used in both the intra-grid of a large organization and in a research grid consisting of large clusters, connected through a high bandwidth dedicated network.

The non-negative matrix factorization toolbox for biological data mining.

BackgroundNon-negative matrix factorization (NMF) has been introduced as an important method for mining biological data. Though there currently exists packages implemented in R and other programming languages, they either provide only a few optimization algorithms or focus on a specific application field. There does not exist a complete NMF package for the bioinformatics community, and in order to perform various data mining tasks on biological data.ResultsWe provide a convenient MATLAB toolbox containing both the implementations of various NMF techniques and a variety of NMF-based data mining approaches for analyzing biological data. Data mining approaches implemented within the toolbox include data clustering and bi-clustering, feature extraction and selection, sample classification, missing values imputation, data visualization, and statistical comparison.ConclusionsA series of analysis such as molecular pattern discovery, biological process identification, dimension reduction, disease prediction, visualization, and statistical comparison can be performed using this toolbox.

A hybrid channel assignment approach using an efficient evolutionary strategy in wireless mobile networks

In wireless mobile communication systems, radio spectrum is a limited resource. However, efficient use of available channels has been shown to improve the system capacity. The role of a channel assignment scheme is to allocate channels to cells or mobiles in such a way as to minimize call blocking or call dropping probabilities, and also to maximize the quality of service. Channel assignment is known to be an NP-hard optimization problem. In this paper, we have developed an evolutionary strategy (ES) which optimizes the channel assignment. The proposed ES approach uses an efficient problem representation as well as an appropriate fitness function. Our paper deals with a novel hybrid channel assignment based scheme called D-ring. Our D-ring method yields a faster running time and simpler objective function. We also propose a novel way of generating the initial population which reduces the number of channels reassignments and therefore yields a faster running time and may generate a possibly better initial parent. We have obtained better results (as well as faster running time) than a similar approach in literature.

Spot Detection and Image Segmentation in DNA Microarray Data

Following the invention of microarrays in 1994, the development and applications of this technology have grown exponentially. The numerous applications of microarray technology include clinical diagnosis and treatment, drug design and discovery, tumour detection, and environmental health research. One of the key issues in the experimental approaches utilising microarrays is to extract quantitative information from the spots, which represent genes in a given experiment. For this process, the initial stages are important and they influence future steps in the analysis. Identifying the spots and separating the background from the foreground is a fundamental problem in DNA microarray data analysis. In this review, we present an overview of state-of-the-art methods for microarray image segmentation. We discuss the foundations of the circle-shaped approach, adaptive shape segmentation, histogram-based methods and the recently introduced clustering-based techniques. We analytically show that clustering-based techniques are equivalent to the one-dimensional, standard k-means clustering algorithm that utilises the Euclidean distance.

Sparse representation approaches for the classification of high-dimensional biological data.

BackgroundHigh-throughput genomic and proteomic data have important applications in medicine including prevention, diagnosis, treatment, and prognosis of diseases, and molecular biology, for example pathway identification. Many of such applications can be formulated to classification and dimension reduction problems in machine learning. There are computationally challenging issues with regards to accurately classifying such data, and which due to dimensionality, noise and redundancy, to name a few. The principle of sparse representation has been applied to analyzing high-dimensional biological data within the frameworks of clustering, classification, and dimension reduction approaches. However, the existing sparse representation methods are inefficient. The kernel extensions are not well addressed either. Moreover, the sparse representation techniques have not been comprehensively studied yet in bioinformatics.ResultsIn this paper, a Bayesian treatment is presented on sparse representations. Various sparse coding and dictionary learning models are discussed. We propose fast parallel active-set optimization algorithm for each model. Kernel versions are devised based on their dimension-free property. These models are applied for classifying high-dimensional biological data.ConclusionsIn our experiment, we compared our models with other methods on both accuracy and computing time. It is shown that our models can achieve satisfactory accuracy, and their performance are very efficient.

Non-negative matrix and tensor factorization based classification of clinical microarray gene expression data

Non-negative information can benefit the analysis of microarray data. This paper investigates the classification performance of non-negative matrix factorization (NMF) over gene-sample data. We also extends it to higher-order version for classification of clinical time-series data represented by tensor. Experiments show that NMF and the higher-order NMF can achieve at least comparable prediction performance.

Classification approach based on non-negative least squares

A non-negative least squares classifier is proposed in this paper for classifying under-complete data. The idea is that unknown samples can be approximated by sparse non-negative linear combinations of few training samples. Based on sparse coefficient vectors representing the training data, a sparse interpreter can then be used to predict the class label. We have devised new sparse methods which can learn data containing missing value, which can be trained on over-complete data, and which also apply to tensor data and to multi-class data. Permutation test shows that our approach requires a small number of training samples to obtain significant accuracy. Statistical comparisons on various data shows that our methods perform as well as support vector machines while being faster. Our approach is very robust to missing values and noise. We also show that with appropriate kernel functions, our methods perform very well on three-dimensional tensor data and run fairly fast.

A new Kernel non-negative matrix factorization and its application in microarray data analysis

Non-negative factorization (NMF) has been a popular machine learning method for analyzing microarray data. Kernel approaches can capture more non-linear discriminative features than linear ones. In this paper, we propose a novel kernel NMF (KNMF) approach for feature extraction and classification of microarray data. Our approach is also generalized to kernel high-order NMF (HONMF). Extensive experiments on eight microarray datasets show that our approach generally outperforms the traditional NMF and existing KNMFs. Preliminary experiment on a high-order microarray data shows that our KHONMF is a promising approach given a suitable kernel function.

Facile whole mitochondrial genome resequencing from nipple aspirate fluid using MitoChip v2.0

BackgroundMutations in the mitochondrial genome (mtgenome) have been associated with many disorders, including breast cancer. Nipple aspirate fluid (NAF) from symptomatic women could potentially serve as a minimally invasive sample for breast cancer screening by detecting somatic mutations in this biofluid. This study is aimed at 1) demonstrating the feasibility of NAF recovery from symptomatic women, 2) examining the feasibility of sequencing the entire mitochondrial genome from NAF samples, 3) cross validation of the Human mitochondrial resequencing array 2.0 (MCv2), and 4) assessing the somatic mtDNA mutation rate in benign breast diseases as a potential tool for monitoring early somatic mutations associated with breast cancer.MethodsNAF and blood were obtained from women with symptomatic benign breast conditions, and we successfully assessed the mutation load in the entire mitochondrial genome of 19 of these women. DNA extracts from NAF were sequenced using the mitochondrial resequencing array MCv2 and by capillary electrophoresis (CE) methods as a quality comparison. Sequencing was performed independently at two institutions and the results compared. The germline mtDNA sequence determined using DNA isolated from the patients blood (control) was compared to the mutations present in cellular mtDNA recovered from patients NAF.ResultsFrom the cohort of 28 women recruited for this study, NAF was successfully recovered from 23 participants (82%). Twenty two (96%) of the women produced fluids from both breasts. Twenty NAF samples and corresponding blood were chosen for this study. Except for one NAF sample, the whole mtgenome was successfully amplified using a single primer pair, or three pairs of overlapping primers. Comparison of MCv2 data from the two institutions demonstrates 99.200% concordance. Moreover, MCv2 data was 99.999% identical to CE sequencing, indicating that MCv2 is a reliable method to rapidly sequence the entire mtgenome. Four NAF samples contained somatic mutations.ConclusionWe have demonstrated that NAF is a suitable material for mtDNA sequence analysis using the rapid and reliable MCv2. Somatic mtDNA mutations present in NAF of women with benign breast diseases could potentially be used as risk factors for progression to breast cancer, but this will require a much larger study with clinical follow up.

STRIP - a strip-based neural-network growth algorithm for learning multiple-valued functions

We consider the problem of synthesizing multiple-valued logic functions by neural networks. A genetic algorithm (GA) which finds the longest strip in V is a subset of K(n) is described. A strip contains points located between two parallel hyperplanes. Repeated application of GA partitions the space V into certain number of strips, each of them corresponding to a hidden unit. We construct two neural networks based on these hidden units and show that they correctly compute the given but arbitrary multiple-valued function. Preliminary experimental results are presented and discussed.