iPromoter-5mC

iPromoter-5mC predicts 5-methylcytosine (5mC) sites within genome-wide DNA promoters to elucidate promoter methylation patterns and their regulatory effects on mRNA gene expression in small cell lung cancer (SCLC).

Key Features:

• CCLE promoter methylation data: Uses promoter methylation datasets from the Cancer Cell Line Encyclopedia (CCLE) with a focus on SCLC cell lines.
• Deep Neural Network (DNN): Employs a deep neural network model to analyze and predict methylation modifications at promoter sites.
• One-Hot Encoding: Encodes promoter samples using One-Hot Encoding as input for the neural network.
• Performance metrics: Reports predictive performance with an average Area Under the Curve (AUC) of 0.957 on independent testing datasets.

Scientific Applications:

• Epigenetic profiling in cancer: Identifies 5mC distribution across promoters to support studies of DNA methylation patterns in cancer biology.
• Methylation–expression analysis: Facilitates investigation of the relationship between promoter methylation and mRNA gene expression regulation.
• Therapeutic target exploration: Supports identification of promoter hypermethylation events relevant to tumor suppressor gene regulation and potential therapeutic strategies in SCLC.

Methodology:

Promoter samples are One-Hot Encoded and analyzed with a deep neural network trained on promoter methylation data from the Cancer Cell Line Encyclopedia (CCLE) focused on SCLC; performance was evaluated using independent testing datasets (average AUC = 0.957).