Wyniki wyszukiwania - BazTech

1

Measuring comparative statistical effectiveness of cancer subtype categorization using gene expression data

Avila Clemenshia P., Deepa C.

Computer Assisted Methods in Engineering and Science

|

2024

|

Vol. 31, no. 2

261--272

EN

This work focused on the analysis of various gene expression-based cancer subtype classification approaches. Correctly classifying cancer subtypes is critical for understanding cancer pathophysiology and effectively treating cancer patients by using gene expression data to categorize cancer subtypes. When dealing with limited samples and high-dimensional biological data, most classifiers may suffer from overfitting and lower precision. The goal of this research is to develop a machine learning (ML) system capable of classifying human cancer subtypes based on gene expression data in cancer cells. These issues can be solved using ML algorithms such as Transductive Support Vector Machines (TSVM), Boosting Cascade Deep Forest (BCD Forest), Enhanced Neural Network Classifier (ENNC), Deep Flexible Neural Forest (DFN Forest), Convolutional Neural Network (CNN), and Cascade Flexible Neural Forest (CFN Forest). In inferring the benefits and rawbacks of these strategies, such as DFN Forest and CFN Forest, the findings are 95%.

2

Tissue Classification Using Efficient Local Fisher Discriminant Analysis

Wang Z., Sun X., Sun L., Qian X.

Przegląd Elektrotechniczny

|

2013

|

R. 89, nr 3b

113--115

EN

A novel scatter-difference-based local Fisher discriminant analysis(SDLFDA) algorithm for tissue classification is proposed in this paper. SDLFDA explicitly considers the local manifold structure and interclass discrimination in gene expression data space. By using SDLFDA, each gene expression data can be projected into a lower-dimensional discriminative feature space. In addition, SDFLDA reduces the computational cost through QR decomposition. Experimental results demonstrate the effectiveness and efficiency of the proposed SDLFDA algorithm.

PL

W artykule przedstawiono algorytm analizy lokalnym wyróżnikiem Fisher’a opartym na różnicach rozproszenia (ang. SDLFDA), służący do klasyfikacji tkanek. Proponowana metoda pozwala na zmniejszenie wymiarowości przestrzeni wyróżnika, określającego dane GXD, a także redukcję kosztów obliczeniowych dzięki dekompozycji QR. Wyniki badań eksperymentalnych potwierdzają skuteczność i sprawność algorytmu.

3

Discovering diagnostic gene targets for early diagnosis of acute GVHD using methods of computational intelligence on gene expression data

Fiasch'e M., Morabito F. C., Verma A., Kasabov N., Cuzzola M., Iacopino P.

Journal of Artificial Intelligence and Soft Computing Research

|

2011

|

Vol. 1, No. 1

81--89

EN

This is an application paper of applying standard methods of computational intelligence to identify diagnostic gene targets and to use them for a successful diagnosis of a medical problem - acute graft-versus-host disease (aGVHD). This is the major complication after allogeneic haematopoietic stem cell transplantation (HSCT) in which functional immune cells of donor, recognize the recipient as ”foreign” and mount an immunologic attack. In this paper we analyzed gene-expression profiles of 47 genes associated with allo-reactivity in 59 patients submitted to HSCT. We have applied different dimensionality reduction techniques of the variable space, combined with different classifiers to detect the aGVHD at onset of clinical signs. This is a preliminary study which utilises both computational and biological evidence for the involvement of a limited number of genes for the diagnosis of aGVHD. Directions for further studies are also outlined in this paper.

4

Application of resampling methods to estimate the number of clusters in gene expression data

Stąpor K.

Theoretical and Applied Informatics

|

2006

|

Vol. 18, nr 2

109-122

EN

In this paper we evaluated the two recently emerged resampling-based methods for estimation the number of clusters (if any) in a dataset. The first method is based on the concept of clustering stability while the second utilizes the ideas from discriminant analysis. These methods are compared using simulated and gene expression data from cancer microarray studies.

PL

W artykule przedstawiono dwie nowe metody dotyczące walidacji grupowania danych, a konkretnie oszacowania liczby grup. Obie metody bazują na odpowiednim próbkowaniu zbioru wejściowego a ich podstawową ideą jest stwierdzenie, że stabilna struktura to taka, która jest odporna" na perturbacje danych. Pierwsza metoda, pochodząca z [1] bazuje na pojęciu tzw. stabilności wyniku grupowania, które z kolei jest definiowane w oparciu o odpowiednio skonstruowaną macierz niezgodności. W drugiej metodzie wykorzystywane są pojęcia z analizy dyskryminacyjnej dla oceny stabilności uzyskanej w wyniku grupowania struktury. Obie metody zostały porównane z użyciem specjalnie wygenerowanych zbiorów testowych. Następnie zastosowano je dla oszacowania liczby grup w danych stanowiących poziomy ekspresji genów pacjentów zdrowych i chorych na różne rodzaje białaczki, pochodzące z mikromacierzy DNA.

5

Selecting Differentially Expressed Genes for Colon Tumor Classification

Fujarewicz K., Wiench M.

International Journal of Applied Mathematics and Computer Science

|

2003

|

Vol. 13, no 3

327-335

EN

DNA microarrays provide a new technique of measuring gene expression, which has attracted a lot of research interest in recent years. It was suggested that gene expression data from microarrays (biochips) can be employed in many biomedical areas, e.g., in cancer classification. Although several, new and existing, methods of classification were tested, a selection of proper (optimal) set of genes, the expressions of which can serve during classification, is still an open problem. Recently we have proposed a new recursive feature replacement (RFR) algorithm for choosing a suboptimal set of genes. The algorithm uses the support vector machines (SVM) technique. In this paper we use the RFR method for finding suboptimal gene subsets for tumor/normal colon tissue classification. The obtained results are compared with the results of applying other methods recently proposed in the literature. The comparison shows that the RFR method is able to find the smallest gene subset (only six genes) that gives no misclassifications in leave-one-out cross-validation for a tumor/normal colon data set. In this sense the RFR algorithm outperforms all other investigated methods.

6

Improved classification of microarray gene expression data using support vector machines

Fujarewicz K., Kimmel M., Rzeszowska-Wolny J., Świerniak A.

Journal of Medical Informatics & Technologies

|

2001

|

Vol. 2, Part. 1

MI9--MI17

EN

Microarrays are new technique of gene expression measurements that attracted a great deal of research interest in recent years. It has been suggested that gene expression data from microarrays (biochips) can be utilized in many biomedical areas, for example in cancer classification. Whereas several, new and existing, methods of classification has been tested, a selection of proper (optimal) set of genes, which expression serves during classification, is still an open problem. In this paper we propose a heuristic method of choosing suboptimal set of genes by using support vector machines (SVMs). Obtained set of genes optimizes one-leave-out cross-validation error. The method is tested on microarray gene expression data of samples of two cancer types: acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL). The results show that quality of classification of selected set of genes is much better than for sets obtained using another methods of feature selection.