MetageneCluster: a Python package for filtering conflicting signal trends in metagene plots

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Metagene plots visualize biological signal trends across genomic subsections.
• They aggregate genome-level data for high-level analysis in sequencing applications.
• Standard metagene plots can mask distinct underlying trends due to signal averaging.

Purpose of the Study:

• To develop a computational tool for generating representative metagene plots.
• To address the limitation of standard metagene plots in handling conflicting signals.
• To enable the identification and summarization of distinct patterns within genomic data.

Main Methods:

• Implementation of MetageneCluster, a Python-based tool.
• Utilizing k-means clustering on genomic regions of interest.
• Clustering data by similarity to identify underlying patterns.

Main Results:

• MetageneCluster generates a collection of representative metagene plots.
• The tool successfully identifies distinct patterns by clustering similar genomic regions.
• Demonstrated effectiveness in uncovering conflicting signals in real-world genomic datasets.

Conclusions:

• MetageneCluster is a user-friendly tool for creating improved metagene plots.
• The tool accurately captures distinct patterns within sequence data.
• Enhances the analysis of complex genomic datasets by revealing hidden trends.