Entering edit mode
24 months ago
catherine
•
0
Hello,
I am currently binning using Metabat2 and I have 5 samples that I have decided to assemble individually as opposed to a coassembly. However, for each assembly I did create sorted bam files for all 5 of my samples reads, since they are from a similar sampling site and I expect some similarities in the MAGs present. However, I am not sure if I should provide Metabat2 with a depth.txt file that I build based on just the one bam file that maps my assembly back to itself, or if I should give it all of the sample bam files even thought the assembly is only built on one sample.
Any advice would be appreciated!
Thank you
Mensur Dlakic - All the samples are from one site at different depths (along a water column). I thought this might provide more information for binning (ie if depth is consistent for a cluster of contigs across all samples and not just one). However, I don't know if metabat would then treat it like a co-assembly by default. It sounds like, from your response, that it is best to only provide one bam file? Thank you for the help!
We have done sampling in two parts of a hot-spring outflow channel that are just a foot apart. There are species that are present in one of those samples but completely absent in the other, and abundance is different even for species that are present in both. In lake sediment samples that are 20 cm apart vertically, a community can go from 70:30 to 40:60 for Bacteria vs. Archaea.
In general, I don't subscribe to using abundance for binning, but that is a topic for a different time (more comments about it are in posts on the right-hand side). If you are using abundance, I don't see any justification for mixing assemblies created for individual sites with reads mapped at multiple sites.