Preventing Overfitting In Transcription Factor Binding Location Prediction Model
Alrashdi, Salma
Alrashdi, Salma
Author
Supervisor
Department
Machine Learning
Embargo End Date
2024-01-01
Type
Thesis
Date
2024
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Forecasting the binding locations of transcription factors is a crucial and expansive field that forms the core of comprehending gene regulatory mechanisms. This interdisciplinary research area merges biology, computational biology, machine learning, and bioinformatics to predict the positions where transcription factors (TFs) interact with DNA sequences, influencing gene expression. The significance of this pursuit is diverse, impacting various biological processes, disease mechanisms, and evolutionary studies. In our model, we intend to adopt a straightforward approach by employing a Convolutional Neural Network (CNN) with Conv1D, a subclass of Conv2D tailored for sequential data. To address the essential need to prevent overfitting in our training data, we plan to incorporate measures such as Dropout layers.
Citation
S. Alrashdi, "Preventing Overfitting In Transcription Factor Binding Location Prediction Model", MS. Thesis, Machine Learning, MBZUAI, Abu Dhabi, UAE, 2024