SequenceServer 1.0 pre-release

sequenceserver
#1

Hey all,

We have been working on a completely new version of SequenceServer. The installation and whole underlying logic and output have been completely rewritten from scratch and there’s tons of new features:

  • Graphical overview of search results per query.
  • Drag and drop a FASTA file to use sequences from a file.
  • View sequence and download FASTA of a single hit, selected hits, or all
    hits (upto 30 hits).
  • Download XML and Tabular report (standard 12 column or all 44 columns).
  • Ability to add custom links for each hit (we will post a howto on this
    soon).
  • Several new options to the command line interface.

We are targeting to push a stable release soon. But before we do so, we would like to hear how you all like the new features and new face of SequenceServer, what we have missed and what we can improve. So we have made a “pre-release”.

To give it a test run:

$ gem install --pre sequenceserver
$ sequenceserver

This will install all necessary dependencies. Please back up your
.sequenceserver.conf file just in case.

We have also setup a demo at http://antgenomes.org/sequenceserver/ but we are expecting a bit of a downtime over the weekend due to scheduled power outage.

Additionally, we have made it 1000x easier than before to get started with
SequenceServer. SequenceServer can now automatically download BLAST+ binaries, ask for the location of FASTA files (or BLAST+ databases), and offer to create BLAST+ databases from FASTA files. “Setup wizard” can be used to configure multiple instances of SequenceServer as well (run ‘sequenceserver -c path/to/where/new/config/should/be/written’).

Our browser compatibility has changed with regards to Internet Explorer (IE). We expect IE10 and above. Some functionality will be broken on IE9 and some more on IE8.

On the server side we are compatible Ruby 1.8.7 - 2.2.0. Just that the pre-rlease version number may create issue with older RubyGems, easily fixed by running:

$ [sudo] gem update --system

We hope you enjoy the new SequenceServer and look forward to hear your feedback.

– Priyam

#2

Nice work guys! Thanks for letting me preview it for a couple of weeks! :slight_smile: I wrote up a blog post on it, along with instructions to run multiple copies at once within Apache - http://guy-leonard.co.uk/setup/sequenceserver-v1-new/ - a little out of date since it was based on pre2 but I’ll update it over the weekend when I upgrade to pre4!

Cheers,

Guy

#3

Just pushed a couple of small fixes to the pre-release:

  • “Full tabular report” download link was not working correctly. Fixed that.
  • 2-3 changes for Ruby 1.8.7 support

To install / update:

$ gem install --pre sequenceserver

– Priyam

#4

Thanks for taking the time to blog about seqserv Guy - we appreciate muchly.

#5

Hello,

Thank you for the update, though I have some questions regarding output results. I tried blasting a random nucleotide sequence on the ant genome server (“cagcagtatgcagtctgatgctgactatgcatgtgacttgctgatgcatgcatgtgactgactggatgctgatgctgatgca”) using Blastx against the Apis dataset ([Outgroup] Apis mellifera proteins prerelease-2). I get 2 results, so far so good.
However, if I then search against 2 databases, (Apis again and Solenopsis invicta proteins 2.2.3) I get 4 results, but I have lost one of the Apis hits (1 Apis, 3 Solenopsis). Furthermore, if I uncheck the Apis database and I am now searching only against Solenopsis, I now have 4 results from Solenopsis and not 3.

My question is, why do you get different numbers of hits when you search against multiple databases?

Thank you.

#6

Hi Adrian,

My suspicion is that this is occurring because of the way e-values are calculated within BLAST - a larger number of sequences in the database (ie selecting >1 db) means that you are more likely to hit a sequence by random chance. E-values are a way to correct for this, by taking the size of the database into account.

Hope that helps.
ben