Which Genome Browser Works Best With Prokaryotic Genomes And Data

Question

[Martin A Hansen]

Do a genome browser exists where you can easily import your own prokaryotic draft genomes (stacks of contigs) and add custom tracks such as NGS mapping data and RAST annotations? I would prefer a web based browser over a desktop based browser, but that totally depends on ease of use and functionality (especially importing custom genomes and tracks).

I find adding custom genomes to Gbrowse painful. Jbrowse is not mature and development has unfortunately stalled. The UCSC Genome browser and Ensembl are not meant for prokaryotes - same for IGV and GenPlay. Artemis don't work well with draft genomes unless you create a single sequence scaffold.

Cheers,

Martin

[girlwithglasses]

FYI JBrowse is a mature genome browser and is now actively developed - seehttp://jbrowse.org for information.

[Martin A Hansen]

I am following JBrowse closely. It is maturing nicely, but still misses the options to setup new genomes and tracks easily.

[girlwithglasses]

Are you on the JBrowse mailing list ( http://jbrowse.org/contact/ )? I suggest that you give your feedback to Rob Buels, the developer; he's extremely receptive to comments and suggestions, and it's the best way to ensure that your concerns get addressed.

[Martin A Hansen]

I know. I have corresponded a bit with Rob.

Answer 1

[Neilfws]

So, you've covered most of the well-known, widely-used genome browsers and you don't like any of them :-)

Perhaps there's something in this quite comprehensive Wikipedia list?

I would give GBrowse another try and be patient, while you work out the best way to use it. The key aspect of GBrowse is to design GFF3 files which are to your liking: i.e. they contain the data that you need, arranged in such a way that a few tweaks to the config file (often a very simple Perl subroutine) will display it as you like. Once you have tried out a few options, it will probably "click" for you and you'll be automating GFF file generation with shell scripts in no time.

A base install of GBrowse is not too difficult these days and personally, I think the flexibility and customization that it offers outweighs the learning curve. I've used it extensively for prokaryotic genomes, with good results.

[Scott Cain]

If you don't mind using the older version of GBrowse (1.70), there was some work to allow circular genomes; that is, you still have the same rectangular panel, but when you scroll off one end, instead of just bumping into the end, you get get the features that are on the "other end" of the GFF file. I don't believe it's perfect, but the developer would probably be interested in suggestions or bug reports. This version of GBrowse can be obtained from a github branch: https://github.com/GMOD/GBrowse/tree/nliles_gbrowse_circular

[Martin A Hansen]

Setting up a new genome in Gbrowse is really painful, while adding custom tracks is merely painful, IMHO. It was clearly developed to display reference genomes and not draft genomes.

[Neilfws]

What's that phrase: no pain, no gain ;-) I disagree that GBrowse is "clearly developed" for reference genomes. It's flexible enough to handle other data, including draft genomes, with a little work.

Answer 2

[Casey Bergman]

Not sure why you discard the UCSC Genome Browser outright, as it has been very successfully adapted by Todd Lowe's lab to warehouse bacterial and archael genomes: http://microbes.ucsc.edu/. They will import base assemblies on request or you can take a page from their book and set up your own UCSC browser instance.

Another option is to use (the main/a local) instance of Galaxy and use its Trackster functionality for visualization of genomic tracks.

[Martin A Hansen]

I have seen this, and I have actually worked a fair bit with a local install the UCSC browser and written scripts for installing new genomes and tracks. But this browser was written for the human genome project and the code still reflects this. I was hoping for something more simple.

Answer 3

[Mary]

How about WebGBrowse? You don't have to do hardly anything....if you are just looking to quickly set something up. http://webgbrowse.cgb.indiana.edu/

[Martin A Hansen]

I looked at WebGBrowse last year and I think there were the same issues with adding new genomes being painful as with the vanilla Gbrowse.

Answer 4

[Konrad]

Another desktop solution that works for microbial genomes, supports various formats and has many features is the Integrated Genome Browser (unfortunately the screenshot on the start page does not look really appealing). Not sure if it is handy to deal with many different contigs but could be worth a try.

[Martin A Hansen]

Looks interesting. Will give it a try.

Answer 5

[Pablacious]

Hi,

I'm surprised that no one had suggested GenDB (from CeBiTeC, Bielefeld University):

http://www.cebitec.uni-bielefeld.de/groups/brf/software/gendb_info/index.html

See guided tour:

http://www.cebitec.uni-bielefeld.de/groups/brf/software/gendb_info/tour.html

Installing it locally is not easy, as you constrained the question, but you can use it over the web (you just need to set up a project with them, is probably a bit of email traffic, and then upload your sequence data). It is built for prokaryotic genomes. We used to use it locally (it has a very good perl API) at my previous lab, but if what you want is reasonable browsing, using it through the web should be fine.

Good luck!

Answer 6

[Val]

Hi Martin (maasha) Have you tried IGB?Did work? Have you ever found any good and easy genome brownser for draft prokarotic genome? VP