Alfresco

The aim is to develop a new visualisation tool that allows effective comparative genome sequence analysis.

The program will compare multiple sequences from putitatively homologous regions in different species. Results from various different existing analysis programs, such as gene prediction, protein homology and regulatory sequence prediction programs shall be visualised and used to find corresponding sequence domains.

The program functions as an stand-alone application and also over the World Wide Web. With this in mind it is being developed using the Java programming language.

A key feature of the program is to use available analysis programs relevant to comparative genome sequence analysis, combine the results of these, and graphically present them in an intuitive way, thereby facilitating the analysis of large genomic regions.We've decided to call this program Alfresco. We originally wanted to call it Fresco (FRont-End for Sequence COmparison) but that name was allready taken.

To run Alfresco you need to have jdk1.1 installed. Alfresco has only been tested on Solaris 2.5, Dec OSF 4.0D and Linux, but should run on any machine with jdk1.1.

New: Alfresco now runs on Mac and Windows as well.

[Genome Research Limited]

Downloads

New to Alfresco?

The manual explains how to install and run the software and what most parts of the program do.

Download Alfresco

  • Download Alfresco, and uncompress and unpack it with zcat alfresco.tar.gz | tar xvf -.

  • Edit the alfresco script to suit your setup, by setting the CLASSPATH variable to point to the alfresco.jar file and setting the path to the java executable. Make the script executable by typing chmod 755 alfresco

    Note: The script has changed from the pre-0.9 versions. Remove your old alfresco script and exchange it with the new one. The change mainly has to do with setting of paths to the external executables, this is now done automagically by Alfreco. Settings for different programs are now stored in a resource file in your home directory.

  • As of version 0.9, CORBA wrappers have been added for several analysis methods. This means that there is not an absolute need to install the programs locally. However, feasability of running them remotely through CORBA depends on your network connection. It will always be faster to use locally installed programs:

    • Dotter;
    • RepeatMasker - Contact Arian Smit (asmit@u.washington.edu) at the University of Washington Genome Center for a copy;
    • Genscan = Contact Chris Burge (cburge@mit.edu) at MIT for a copy;
    • cpg. Binaries (OSF, Solaris and Linux) and source code are included in the cpg.d directory of the distribution;
    • DBA and genewise are part of Ewan Birneys Wise2 package. A modified blastwise scripts is available in the blastwise directory (see 'blast database searching' below).
    • est_genome. A script, blEst_genome, for combining blast searches with est_genome is available in the blEst_genome directory (see 'blast database searching' below).
    • For doing blastn-alignments you need:
    • For blast database searching you need:
      • Blast 2 (again);
      • Local copies of suitable databases e.g. EMBL, Swissprot and TREMBL. Also, it's good to have subsets of EMBL, like the EST division, and the RNA entries excluding EST sequences;
      • SRS for the blastwise and blEst_genome scripts. Or, some other way to retrieve the matching entries from your databases (you have to modify the blastwise and blEst_genome scripts).
      • bioperl modules for the blastwise and blEst_genome scripts.
    • Make sure to install all programs in directories that are in your PATH variable!
  • If you want to use a CORBA server, the only server currently running is available through the IOR available at http://kisac.cgr.ki.se/nic/methods_server.ior. This url is set by default by Alfresco.

  • In the future I would like others to set up CORBA servers to increase the choice of servers for the users. If you want to set up your own CORBA server please contact me and I'll help you.

  • When you are all set, just type alfresco

To have something to look at I've included two cmp files and the fasta sequence files that go with them. Choose "Read .cmp file" from the File Menu to open them, or type alfresco <cmp file> on the command line.

Anyone interested in improving the source code is most welcome. If you do, Niclas would appreciate if you'd send him the updates.

Contact

If you have any problems, please don't hesitate to contact: Niclas Jareborg

Screenshots

The display area shows representations of two, approximately 90 kb long, EMBL sequence entries of the corresponding regions from mouse and human. Genes are represented by blue borders and a name tag, with ">" and "<" indicating the direction of transcription. Genes encoded by the sense strand are drawn above the line representing the sequence, and genes encoded by the antisense strand are drawn below. Coding and non-coding exons are indicated in green and yellow, respectively. Gray lines connect homologous exons. Repeats are indicated by smaller boxes in blue, green, brown, and pink depending on type of repeat. Predicted CpG islands are shown as yellow boxes one level away from the line representing the sequence.

The sequences are independently scrollable and zoomable.

Screenshot Screenshot Screenshot Screenshot Screenshot
* quick link - http://q.sanger.ac.uk/66rst3d0