How to use RefineM to Improve completeness

[nohayoussef]

Hello,
I was wondering how we can use RefineM to
Improve completeness of already binned MAGs, specifically:

1. identify contigs with similarity to specific reference genome(s)
2. identify contigs with compatible GC, coverage, and tetranucleotide signatures
3. identify partial MAGs which should be merged together (requires CheckM)

All of these are stated on RefineM help page but I am not sure exactly how to implement them and what input files are needed for these steps. From glancing over the refinem page, I learned that I need to use the commands reference and compatible? Is this true? and if so, what do I need as input.
Thank you so much for your time and for a great tool!

[donovan-h-parks]

Hello. I haven't work out a solid workflow for doing this yet. The idea is as you describe, but exactly how this would be automated and how best to ensure chimeric MAGs aren't produced is still an open question. If you explore this possibility I'd be very interested to know the approach you took and the degree to which it improved your MAGs.