How to decide on a reference genome

[User1234567]

Hi,

We are trying to assemble an unknown bacterial genome from Ion Torrent data.

So far we have assembled contigs using mira. We tried a few gap-filling tools but didn't get very far with that.

We now want to find the best reference genome possible for our sequence. Does anyone have advice on how we can do this?

[Henri.]

Well, you could take a long contig and BLAST it, or BLAST all the contigs and see what comes up.
When you filter out a certain bacteria make sure you map your reads to all available strains because you dont know if you sequenced a new strain or which of the existing ones.