Summary:
Barr Ticknor asks if a comprehensive package of biocomputing software can be
assembled for free or cheaply or remotely access for a small fee.
I believe if your users are not fussy, it is possible to survive on a
really low or no budget facility. You can get all the software listed below
for free at the usual ftp sites and email server sites or contact me.
Getting nice pictures for publications may be a slight problem, as you
will have to do some computing gymnastics (well, they're free after all)
and play around.
Off the top of my head...
a) For general sequence analysis, for pcs, use SEQAID (Doug Rhoads et al)
for macs, use Seqapps (from Don Gilbert)
They are all very nice software, and more importantly, they're free.
SEQAID for instance does simple alignment, with or without gaps,
translation, backtranslation, simple sec struct/antigenicity prediction,
look for restriction sites etc.
b) On the pc, you can use Clustalv for multiple sequence alignment,
Readseq for interchanging sequence file format, Profilegraph for
protein sequence analysis, prosearch for identifying PROSITE motifs,
ELBAMAP for electrophoresis band mapping, COMAP for restriction map
construction, PHYLIP phylogenetic inference, SIGSEQ signal peptide
prediction, etc. all for free.
c) On the unix, use clustalv (much quicker than on pc), readseq, prosearch,
phylip, PIMA, GDE, STADEN, FSAP, BLOCKS, PRIMER (pcr primer
prediction) etc.
d) For sequence retrieval, it is definitely much faster by gopher and wais
(even from Singapore) than using GCG on our unix machine.
More than two dozen major molecular biology databases are now available
freely,
Genbank, Genpept, EMBL, DDBJ, GDB, PDB, PROSITE, TFD, VECTOR, ECD,
ENZYME, REBASE, SWISSPROT, PIR, EST, KABAT, FLYBASE, ACEDB, AATDB,
ECOLI, RDP (RNA),
e) BLAST, GRAIL, GENEID and other email servers, (use MAILFASTA) can fill in
the gaps.
We've got the PC software integrated under a menu interface on our
local area network, and the Unix software including GOPHER and WAIS
retrievals under our HYBROW (see CABIOS 9(5)) interface.
GCG is only an icing on the cake for us. I believe if one has the time
to organise things on your machine, it is easily possible to live without GCG,
not that GCG is that expensive to maintain.
We collaborate with
Institut Pasteur (tekaia at pasteur.fr)
Fred Hutchinson Cancer Research Center (mparker at fred.fhcrc.org)
World Data Center, RIKEN, Japan(sugawara at viola.riken.go.jp),
in providing friendly user interfaces for biocomputing.
Ask them too.
TW "it's a great network" Tan
Tin-Wee TAN / INTERNET: bchtantw at nuscc.nus.sg
Biocomputing Research and User Support (BRUS) and
Dept of Biochemistry / BITNET: BCHTANTW at NUSVM
National University of Singapore / Tel: (65)-772-3678
Singapore 0511 / Fax: (65)-779-1453
On 2 Sep 1993, Reinhard Doelz wrote:
> In article <2635egINNt98 at newsstand.cit.cornell.edu>, Barr Ticknor <bet1 at cornell.edu> writes:
> |> The Cornell Biotechnology Program currently maintains the GCG package
> |> and accompanying databases as the core of its biotechnology computing
> |> facility. Because this facility is operating at a loss, there is a very
> |> real possibility that it will cease to exist. What are the alternatives
> |> for researchers who want the functions provided by GCG but don't have
> |> access to a local site? Are such services available through the network,
> |> either freely or on a fee per use basis? I am aware that BLAST searches
> |> are provided by the NCBI server, and that sequence records can be
> |> retrieved from both NCBI and EMBL, but what about the rest? What do
> |> people do at institutions that don't have the GCG package?
>> There's literally no alternative to access a _comprehensive_ package
> on a reasonable computer. You can go easily for the PC or Mac track
> if you have only one or two hardcore users but a full site will not
> operate economically in buying Licenses for too many PC's or Macs
> (as you seem to have financial constrains). As you need more than just
> search and retrieve I would try to make a point to your financing department
> that if you do not provide the services locally you must BUY these services
> in terms of access time to a computer where you can achieve an agreement to
> purchase access time. In Europe, the European Molecular Biology Network
> provides national services which do precisely this but as you are in the US
> you would need to ask around. As the support is also remote, then,
> possibly significant cost arise from losses due to educational issues.
>> Regards
> Reinhard
