What is it about?
The article presents a new clustering approach, voomSOM, that combines voom and self-organizing maps with k-means, k-medoid, and hierarchical clustering algorithms to cluster RNA-seq data. The voom method transforms the RNA-seq count data into a log-cpm matrix, and the SOM algorithm generates a codebook used in downstream analysis. The approach is evaluated on simulated and real datasets and performs similarly or better than other methods.
Photo by Sangharsh Lohakare on Unsplash
Why is it important?
The integration of voom and SOM enhances the clustering algorithms' performance in overdispersed RNA-seq data. The proposed voomSOM approach is an efficient and novel clustering method for RNA-seq data.
Read the Original
This page is a summary of: voomSOM: voom-based Self-Organizing Maps for Clustering RNASequencing
Data, Current Bioinformatics, February 2023, Bentham Science Publishers, DOI: 10.2174/1574893618666221205154712.
You can read the full text:
The following have contributed to this page