Mobile elements are genomic elements exchanged outside of vertical inheritance mechanisms. They include, but are not limited to, plasmids, transposons, bacteriophages, repeat regions, and genomic islands, and may be exchanged within a species (lateral), between different species (horizontal), or across different kingdoms (orthogonal). We are interested in the role that mobile elements play in ecological niche colonization by microbes, especially the development of endosymbioses between bacteria and eukaryotic hosts.

Wednesday, October 22, 2014

The surprisingly vacant world of usable tools to predict prokaryotic promoters

A few days ago, I needed to find (predicted) promoter regions in a particular closed bacterial genome. Naively, I thought I'd just waltz into the Internet for Bioinformaticians-Who-Just-Want-To-Get-Shit-Done, find several-to-many tools to accomplish the task, and waltz out.


Ah, silly me.

BDGP. DNF. Berkley Drosophila Genome Project. "This server runs the 1999 NNPP version 2.2 (March 1999) of the promoter predictor." Um... Also, eukaryotic.

BPROM. FAIL. My full genome file caused BPROM to return a server error. Tried to run 1/5 of the genome and it hung for 5 minutes before I gave up.

ClassProm. FAIL. Interface available, will accept a file upload, results no longer available.

Dragon Genome Explorer. Link Rot.

PePPER. FAIL. Pasted in my genome file and gene coordinates. Run started and immediately reset to the landing page.

PromAn. Link Rot.

PromBase. OK. Fortunately, my genome is in their database of pre-calculated promoters. Unfortunately, I have to download the entire (500mb) database, and apparently it is hosted on a very slow server (ETA 1-4 hours from my university connection).

PromScan. FAIL. Submitted my genome file, ended up at a 404 Error.

PPP. OK. Requires intergenic region sequences extracted from the genome. Actually yielded a results file, although the data format is horrific.

Virtual Footprint. FAIL. Unable to configure the input settings properly. Seems to require a separate run for each gene of interest, problematic for analyzing an entire genome.

There were a few additional tools that apparently no longer exist, and it would not surprise me if I've missed others. But so far, I'm batting about 150, and not a good 150 either. I'm now faced with a one-off scrape of the PromBase database, or a hefty amount of very specific file conversion scripting for PPP.

Either way, it really cuts into my time saving the world from marauding microbial mobile elements.


Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)