GraphBin-Tk: assembly graph-based metagenomic binning toolkit

The study of genetic material directly obtained from natural environments, termed metagenomics, offers valuable insights into microbial communities and their impact on human health and environmental dynamics. Once the genetic material is extracted, sequenced to obtain reads, and assembled to obtain contigs, a process known as metagenomic binning is used to cluster contigs into bins that represent different taxonomic groups, which results in draft microbial genomes or metagenome-assembled genomes (MAGs). Several automated metagenomic binning tools incorporating novel computational methods have been introduced which have led to the discovery and characterisation of many novel micro-organisms.

Conventional metagenomic binning tools make use of features such as nucleotide composition and abundance information of contigs, yet find it challenging to bin closely related species and contigs with noisy features. Binning tools, such as MetaCoAG that use assembly graphs (which contain the connectivity information) are gaining popularity due to their improved binning results over conventional binning methods. Moreover, assembly graph-based bin refinement tools such as GraphBin and GraphBin2 have been introduced to refine binning results from existing binning tools. Yet, these tools exist as individual software packages and running them individually can be complex, time-consuming, and less accessible. Here we present GraphBin-Tk, an assembly graph-based metagenomic binning tool that combines the capabilities of MetaCoAG, GraphBin and GraphBin2, along with additional pre-processing and post-processing functionalities into one comprehensive toolkit. GraphBin-Tk is hosted at https://github.com/metagentools/gbintk.