Utilizing deep learning models for the identification of enhancers and super-enhancers based on genomic and epigenomic features

Super-enhancers are a category of active super-enhancers densely occupied by transcription factors and chromatin regulators, controlling the expression of disease-related genes and cellular identity. Recent studies have demonstrated the formation of complex structures by various factors and super-en...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of intelligent & fuzzy systems 2024-03, p.1-11
Hauptverfasser: Ahani, Zahra, Shahiki Tash, Moein, Ledo Mezquita, Yoel, Angel, Jason
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Super-enhancers are a category of active super-enhancers densely occupied by transcription factors and chromatin regulators, controlling the expression of disease-related genes and cellular identity. Recent studies have demonstrated the formation of complex structures by various factors and super-enhancers, particularly in various cancers. However, our current knowledge of super-enhancers, such as their genomic locations, interaction with factors, functions, and distinction from other super-enhancers regions, remains limited. This research aims to employ deep learning techniques to detect and differentiate between super-enhancers and enhancers based on genomic and epigenomic features and compare the accuracy of the results with other machine learning methods In this study, in addition to evaluating algorithms, we trained a set of genomic and epigenomic features using a deep learning algorithm and the Python-based cross-platform software to detect super-enhancers in DNA sequences. We successfully predicted the presence of super-enhancers in the sequences with higher accuracy and precision.
ISSN:1064-1246
1875-8967
DOI:10.3233/JIFS-219356