using PhyloSift to analyze metagenome-assembled genomes

[Zhou Li]

Hi PhyloSift developers, 

I am a first time user of PhyloSift and I am also new to phylogenomic analysis, so I may ask some naive questions. 

I have dozens of metagenome-assembled genomes (MAGs) and a few hundred of reference genomes. My goal is 1) infer taxonomy of those dozens of MAGs; 2) build a phylogenetic tree that contains dozens of MAGs and a few hundreds of reference genomes. I am hoping PhyloSift can help me achieve my goals. 

I believe I have gotten the PhyloSift working on my computer because I didn't see any error message when I ran phylosift using two of my MAG as input (a bunch of contigs from metagenome assembly in a fasta file) and I have seen many output files. 

My questions are: 

1: if I want to know the taxonomy of MAG, am I correct that I should go to taxa_90pct_HPD.txt and find the one with the highest probability mass? 

2: to make a phylogenetic tree that contains MAGs and a few hundreds of reference genomes, how should I run phylosift to achieve that? Should I run phylosift for each genome separately or should I run all those genomes in one batch? If I run each genome separately, would I automatically get one concatenated alignment from phylosift for ALL of those genomes to make a tree or I would get a concatenated alignment for each genome separately and then I would have to concatenate those alignments myself? 

Thank you for taking time to answer my question.

Zhou