METACLUSTERplus - an R package for probabilistic inference and visualization of context-specific transcriptional regulation of biosynthetic gene clusters

Abstract

Fungi and plants reveal widespread occurrences of metabolic enzymes co-located on the chromosome, some already characterized as being biosynthetic pathways for specialized metabolites, such as terpenes synthesizing enzyme clusters in Lotus japonicus and Arabidopsis thaliana. These clusters display context-specific co-expression of clustered enzymes, indicating a shared transcriptional response in a spatial and condition specific manner, and co-regulation due to promoter binding by shared transcription factors may be one way to facilitate coordinated expression. To enhance our understanding of context-specific transcriptional gene cluster regulation, we redefine and augment this probabilistic framework, labelled METACLUSTERplus, integrating gene expression compendia, context-specific annotations, biosynthetic gene cluster definitions, as well as gene regulatory network architectures. Further, it provides a set of appealing and intuitive visualizations of inferred results for analysis and publication. METACLUSTERplus is available at https://github.com/mbanf/MetaclusterPlus.