Wyniki wyszukiwania - BazTech

1

MicroRNA expression prediction: Regression from regulatory elements

Oğul H., Tuncer M. E.

Biocybernetics and Biomedical Engineering

|

2016

|

Vol. 36, no. 1

89--94

EN

MicroRNAs are known as important actors in post-transcriptional regulation and relevant biological processes. Their expression levels do not only provide information about their own activities but also implicitly explain the behaviors of their targets, thus, in turn, the circuitry of underlying gene regulatory network. In this study, we consider the problem of estimating the expression of a newly discovered microRNA with known promoter sequence in a certain condition where the expression values of some known microRNAs are available. To this end, we offer a regression model to be learnt from the expression levels of other microRNAs obtained through a microarray experiment. To our knowledge, this is the first study that evaluates the predictability of microRNA expression from the regulatory elements found in its promoter sequence. The results obtained through the experiments on real microarray data justify the applicability of the framework in practice.

2

Genomic Virtual Laboratory

Cegielska B., Kaliszan D., Handschuh L., Figlerowicz M., Meyer N.

Computational Methods in Science and Technology

|

2010

|

Vol. 16, No. 1

39-49

EN

In contemporary science, virtual laboratories give a chance to improve research by facilitating access to high-throughput technologies and bioinformatics methods. The Genomic Virtual Laboratory (GVL) presented here was developed for automate analysis of data retrieved from a microarray experiment. The system was implemented for R Bioconductor-based analysis of results obtained in the study on human acute myeloid leukaemia (AML). The article extends the theoretical aspects of GVL presented earlier [8] and describes how the particular elements were integrated to establish the advanced system of two-colour microarray data analysis.

3

An Extension of TSP-family Algorithms for Microarray Classification

Czajkowski M., Krętowski M.

Zeszyty Naukowe Politechniki Białostockiej. Informatyka

|

2009

|

Z. 4

31-45

EN

Classification of microarray data and generation of simple and efficient decision rules may be successfully performed with Top Scoring Pair algorithms. TSP-family methods are based on pairwise comparisons of gene expression values. This paper presents a new method, referred as Linked TSP that extends previous approaches kˇTSP and Weight kˇTSP algorithms by linking top pairwise mRNA comparisons of gene expressions in different classes. Opposite to existing TSP-family classifiers, the proposed approach creates decision rules involving single genes that most frequently appeared in top scoring pairs. Motivation of this paper is to improve classification accuracy results and to extract simple, readily interpretable rules providing biological insight as to how classification is performed. Experimental validation was performed on several human microarray datasets and obtained results are promising.

PL

Klasyfikacja danych mikromacierzowych a także późniejsza interpretacja reguł decyzyjnych może być skutecznie przeprowadzona za pomocą metod z rodziny Top Scoring Pair, polegających na analizie par genow o przeciwstawych poziomach ekspresji w róźnych klasach. W poniższym artykule zaprezentowano nową metodę: Linked TSP, ktora rozszerza działanie klasyfikatorów k-TSP i Weight k-TSP. W przeciwieństwie do algorytmow z rodziny TSP proponowane rozwiązanie tworzy reguły decyzyjne zbudowane z pojedynczych genów, co znacznie ułatwia ich późniejszą interpretację medyczną. W algorytmie wykorzystywane są pary genow uzyskane z algorytmow TSP z których następnie, wybierane są pojedyncze, najczęściej powtarzające się geny. Testy algorytmu Linked TSP przeprowadzone zostająy na rzeczywistych zbiorach danych pacjentow a uzyskane wyniki są obiecujące.

4

Decision tree approach to microarray data analysis

Grześ M., Krętowski M.

Biocybernetics and Biomedical Engineering

|

2007

|

Vol. 27, no. 3

29-42

EN

The classification of gene expression data is still new, difficult and also an interesting field of endeavour. There is a demand for powerful approaches to this problem, which is one of the ultimate goals of modern biological research. Two different techniques for inducing decision trees are discussed and evaluated on well-known and publicly available gene expression datasets. Empirical results are presented.

5

Application of the adaptive noise removal technique to the enhancement of cDNA microarray images

Smolka B., Bieda R.

Journal of Medical Informatics & Technologies

|

2005

|

Vol. 9

131--142

EN

In this paper a novel class of filters designed for the removal of impulsive noise in colour images is presented. The proposed filter family is based on the kernel function which controls the noise suppression properties of the new filtering scheme. The comparison of the new filtering method with the standard techniques used for impulsive noise removal indicates its superior noise removal capabilities and excellent structure preserving properties. The proposed filtering scheme has been successfully applied to the denoising of the cDNA microarray images. Experimental results proved that the new filter is capable of removing efficiently the impulses present in multichannel images, while preserving their textural features.