Hello,

I am interested in the selection statistics (pi, Tajima’s D) of a few Gardnerella vaginalis genes. In order to best calculate these stats I want as many G. vaginalis genomes as possible. So, I have generated G.vaginalis MAGs from vaginal metagenomes (https://virgo.igs.umaryland.edu/).

The next step is where I am unsure.

My goal is to create individual alignments of my genes of interest from my MAGs.

Example:

geneA.fasta

>MAG1
actgactg
>MAG2
actaactg
>MAG3
actgactg

A challenge with Gardnerella is that the genes I’m interested are not universally annotated. Some annotators call them as hypothetical or as different names.

Do you know of good tools/approaches to make gene alignments from metagenomes?

My first thought was to use Roary which is a core genome aligner & gene predictor. https://sangerpathogens.github.io/Roary/ I can feed it annotated gff files from a prokka annotation of each of my MAGs, and it will generate alignments for each gene.

Another idea I had is to somehow align the MAGs to a reference, but I’m not sure if that would work.