Genome bins comming off automated pipelines can be contaminated with parts of other genomes. As part of my workflow I use CheckM (I’m biased since I’m a coauthor) to assess the contamination of genome bins using single-copy marker genes. If you’re lucky then the genome bins that you’re interested in will be relatively complete without much contamination. Unfortunately that isn’t always the case. In this blog post I’m going to run through some of the analyses that I did on a genome bin that was 90% complete but 70% contaminated. This is exploratory analysis to see if I can manually improve the bin over what the automated tools can do.