JMnorm: a novel joint multi-feature normalization method for integrative and comparative epigenomics.

Area of Science:

• Epigenomics
• Genomic Regulation
• Computational Biology

Background:

• Epigenetic features combinatorial patterns reflect genomic region states and functions.
• Existing normalization methods often analyze epigenetic features independently, potentially distorting biological relationships.
• This limitation hinders comprehensive biological interpretation of epigenomic data.

Purpose of the Study:

• To introduce JMnorm, a novel approach for simultaneous normalization of multiple epigenetic features.
• To leverage correlations between epigenetic features for improved data normalization.
• To enhance the biological interpretability and utility of epigenomic datasets.

Main Methods:

• JMnorm simultaneously normalizes multiple epigenetic features by utilizing information from partially correlated features.
• The approach is designed to be applicable across different cell types, species, and experimental conditions.
• Comparative analysis against existing normalization methods was performed.

Main Results:

• JMnorm-normalized data better preserve cross-epigenetic-feature correlations across cell types compared to other methods.
• JMnorm enhances consistency between biological replicates.
• JMnorm consistently improves performance in downstream analyses like clustering, prediction, and transcription factor binding detection.

Conclusions:

• JMnorm effectively minimizes technical noise while preserving true biological relationships within epigenetic datasets.
• The method enhances integrative and comparative epigenomics by improving data quality and interpretability.
• JMnorm offers a robust solution for analyzing complex epigenomic data.