Multi-sample binning using fairy coverage

[Louis-MG]

Hello ! I would like to use fairy for coverage estimation prior running semibin2.

I would like to use the multi-sample approach, is it possible with the AEMB file format ? The documentation says all against all, but multi-sample binning requires all samples to be aligned against a concatenated fasta of scaffolds. It feels similar.

Are the two methods compatible and which instructions do I use ?

Have a nice week-end.

[jolespin]

@BigDataBiology I'm also interested in this. I have a few questions in this vein:

1. I understand how to use the --depth-metabat2 argument and that -a can only be used with strobealign with 5 or more samples but I'm still a little confused on what to do if you have 2-4 samples in a co-assembly?

2. here's the docs --depth-metabat2: depth file generated by metabat2 (only used with single-sample binning).. If I have a metabat2 coverage file created from CoverM with 2+ samples, will this fail?

3. what is going on with the strobealign coverage file creation? There's a split command but I'm not sure what is happening exactly here and why you need to split before strobealign. Can anyone enlighten me on this?

4. If multi_easy_bin is analogous to the multi-split binning in VAMB or VEBA, how can you do co-assembly binning w/ 2-4 samples?