E-Locus - Institutional Repository of the University of Crete

Home Collections Type of Work Post-graduate theses

Post-graduate theses

Current Record: 11 of 6663

[Add to Basket]

Identifier

000466679

Title

Optimization of γ-OMP algorithm for feature selection in gene expression data

Alternative Title

Βελτιστοποίηση του αλγορίθμου Γ-OMP για την επιλογή χαρακτηριστικών σε δεδομένα γονιδιακής έκφρασης

Author

Τζαγκαράκη, Σοφία

Thesis advisor

Τσαγρής, Μιχαήλ

Reviewer

Νικολάου, Χριστόφορος
Πανταζής, Ιωάννης

Abstract

The advent of next-generation technologies revolutionized the field of biotechnology, allowing scientists to sequence hundreds of thousands of genes or an entire gene in a short amount of time. It can also focus on mapping variants/mutations that may have biological signaling implications for normal cell function. As a result, these findings have significant implications for illness diagnosis, prognosis, therapy decisions, and patient monitoring. This groundbreaking technological possibility of massively parallel sequencing opens up new opportunities for precision medicine, but it also introduces a new challenge: data overload. First and foremost, genomics examines massive volumes of data, frequently including hundreds of genes in numerous samples. Identifying the proper selection of genes for identifying biological states is critical to understanding the underlying biological mechanisms. In this study, we looked at the most appropriate feature selection in cancer-related gene expression data. We developed a feature selection method (dcor-OMP), a version of the γ-OMP algorithm. The performance compared to the γ-OMP precursor was also assessed. Both of these methods are variations on the Orthogonal Matching Pursuit (OMP), which is widely used in signal processing (Chen, Billings, & Luo, 1989; Davis, Mallat, & Zhang, 1994), and have been specially customized for gene expression analysis

Language

English, Greek

Subject

Distance correlation

Microarray technology

Next -generation sequencing

Αλληλουχία επόμενης γενιάς

Μηχανική μάθηση

Τεχνολογία μικροσυστοιχιών

Issue date

2024-07-26

Collection