Back to search

University of Patras

Computational methods for transcriptomic data analysis

Abstract

dc:description

The rapid advancement of next-generation sequencing (NGS) technologies, coupled with sophisticated computational techniques, has revolutionized our understanding of short non-coding RNAs and their roles in cellular processes and disease. In this work, we develop computational machine learning based methods to analyze RNA-Seq data and also unravel the complex landscape of non-coding RNAs. Our computational methods leverage previously unaligned reads to identify both known and predicted ncRNAs, offering deep insights into their functional roles. Importantly, our research demonstrates that our methods outperform existing tools in terms of predictive performance, enabling more reliable analysis of RNA-Seq data and enhancing our understanding of non-coding RNAs in disease. Using such transcriptomics data and machine learning methods, we develop predictive models for diagnosis and prognosis. Cutaneous melanoma is selected as a use case. Thus first we address the challenge of fragmented data in cutaneous melanoma research by creating the ebioMelDB database, which integrates diverse clinical, imaging, and RNA-Seq data. This user-friendly resource empowers researchers to develop more accurate diagnostic tools, prognostic models, and therapeutic strategies. We subsequently utilize the data from ebioMelDB to develop targeted predictive models for cutaneous melanoma diagnosis, and prognosis. Our machine learning-based classifiers, incorporating gene expression data and miRNA profiles, successfully predict vital status, recurrence, and metastasis in cutaneous melanoma with high accuracy. Furthermore, we explore the potential of deep learning in skin cancer diagnosis using both dermoscopy and clinical images. By focusing on publicly available datasets, we develop a deep learning classifier capable of discriminating skin cancer from benign lesions. Overall, this work showcases the potential of machine learning techniques, advanced bioinformatics tools, and comprehensive data resources to enhance the understanding, diagnosis, and treatment of cutaneous melanoma.

Degree

thesis:*
Grantor dc:publisher
University of Patras
Year dc:date
2024

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Korfiati, Aigli
  • Κορφιάτη, Αίγλη

Subjects

dc:subject × 20

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/57882

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Korfiati, Aigli; Κορφιάτη, Αίγλη. Computational methods for transcriptomic data analysis. University of Patras, 2024. http://hdl.handle.net/10442/hedi/57882