Homologous Invertebrate Genes Database
HOINVGEN release 01 Online on March 2005
Institute for Computer Sciences
Department of Bioinformatics
You can query Hoinvgen using
Quick Search or
Alternatively you can use the following links:
HOINVGEN is a database of homologous invertebrate genes, structured under
ACNUC sequence database management system.
It allows one to select sets of homologous genes among invertebrate species,
and to visualize multiple alignments and phylogenetic trees. Thus HOINVGEN
is particularly useful for comparative sequence analysis, phylogeny and
molecular evolution studies. More generally, HOINVGEN gives an overall view
of what is known about a peculiar gene family.
The version of HOINVGEN
HOINVGEN is build in the same way as HOBACGEN (for more details, click
here, or see the paper on HOBACGEN by Perrière et al. Genome Res 2000 10:379-85).
Secondly, we developped a new graphical interface (named FamFetch). This interface is written
in JAVA and should work on any computer (Mac, PC, UNIX, etc.), and does not require to have the whole database installed locally.
The database itself contains all invertebrate protein sequences from UniProt (SWISS-PROT+TrEMBL),
with some data corrected, clarified or completed (notably to address the problem of redundancy and
orthology/paralogy) and with some annotation modifications.
It contains also all the corresponding nucleotide sequences in EMBL.
Homologous proteins are classified into families and multiple alignments and phylogenetic trees
are computed for each family.
Sequences and related information have been structured in an ACNUC database.
The description on how the database is built is available
The present version of HOINVGEN is release 01 (March 2005). It has been built
using sequences from UniProt 2.5 (SWISS-PROT 44.5 and TrEMBL 27.5) (September 2004). It
contains a total of 132,023 protein sequences (and 116,498 nucleic sequences) classified in
Among all the proteins included in this release, 96,352 (73%) are classified
into 10,073 families containing at least two sequences, and
35,672 (27%) partial proteins are not attached to a family.
Graphical User Interface
HOINVGEN interface is based on a client/server architecture. To access
the database you only need to install the FamFetch
application on your computer. This program, written in Java, integrates a GUI that allows
users to easily access and visualize:
In FamFetch phylogenetic trees, genes are colored according to the
species from which they come. The user can modify the color table
according to the taxa (any taxonomic level) he is interested in. This
color table is saved in a file of preferences (named .famfetch in
UNIX (UNIX, Linux, Mac OS X), FamFetch.Prefs in Mac OS 9 (or earlier), FamFetch.ini in Windows systems). The color
table that is installed by default with FamFetch is dedicated to
prokaryotes (for the HOBACGEN database). You can replace this preference file by the one we have prepared for invertebrates, that is available here.
- The list of the families available in the database.
- The sequence (protein or nucleotide) of the genes
defining these families.
- The alignments built with these families.
- The phylogenetic trees computed with these alignments.
It is also possible to query the database on this server through the
WWW-Query system. Note that
HOINVGEN is splitted into two databases on this server: HOINVPROT
contains the protein sequences from SWISS-PROT + TrEMBL while HOINVNUCL
contains the nucleotide sequences from EMBL.
You don't need to install the server itself to have HOINVGEN running on your
computer as the client is enough for that purpose. On the other hand
you may want to set-up your own server in a way to speed up your
database access and to propose that service to potential users in your
geographic area. To install an HOINVGEN server, you need first to
register. Starting from the
registering page results, you will have access to the server installation
The whole database is available from our FTP server at URL:
Note that it is much more efficient to use a dedicated FTP client to download the
database rather than an Internet Web browser.
Important note: the SWISS-PROT entries such as those found in HOINVGEN are
copyrighted. They are produced through a collaboration between the Swiss Institute
of Bioinformatics and the European Bioinformatics Institute. There are no restrictions
on its use by non-profit institutions as long as its content is in no way modified.
Usage by and for commercial entities requires a
(See or send an Email to firstname.lastname@example.org).
Contact and reference
If you encounter some problems when installing or using HOINVGEN, please
contact Ingo Paulsen or
Also we welcome any comments or suggestions on the database and/or its
If you have problems or comments...
Back to PBIL home page