Mail server for protein domain funct. homologies - ICGEB Trieste

Valeria Bevilacqua valeria at genes.icgeb.trieste.it
Fri Dec 10 03:55:35 EST 1993


        SS   BBB      A      SS   EEEE      H   H  EEEE L    PPP  
       S  S  B  B    A A    S  S  E         H   H  E    L    P  P 
       S     B B    A   A   S     E         H   H  E    L    P  P 
        S    BB     AAAAA    S    EEE       HHHHH  EEE  L    PPP  
         S   B B    A   A     S   E         H   H  E    L    P    
          S  B  B   A   A      S  E         H   H  E    L    P    
      S   S  B  B   A   A  S   S  E         H   H  E    L    P    
       SSS   BBB    A   A   SSS   EEEE      H   H  EEEE LLLL P    

              ----------------------------------
     This is the help file of the SBASE Email Server at the
International Centre for Genetic Engineering and Biotechnology
     AREA Science Park, Padriciano 99, 34012 Trieste, Italy.
              ----------------------------------

The SBASE Email Server is at present experimental. Please send comments 
          to sbase-comment at icgeb.trieste.it. Thanks.


The SBASE e-mail server accepts a specially formatted mail message
containing a protein query sequence, and, as a response, it sends the list
of the most probable domain homologies. A database search is performed 
against the SBASE library of protein domains using the BLAST algorithm, 
and the search results, provided with annotations, are returned in a mail 
message.


How to use:

Getting HELP:
============

This help file can be gotten by sending an email to

    	    sbase at icgeb.trieste.it

containing the word HELP alone in the first line of the message body.

Making a QUERY:
===============

Send e-mail to sbase at icgeb.trieste.it using the below example

Note: the parameters before the token BEGIN are optional, (here the 
defaults are listed), the lines after BEGIN are required.

Example:

MATRIX PAM120
SCORE PARAMETER 35
ANNOTATIONS YES
BEGIN
> mysequence
LRNGVDINTCNQNGLNGLHLASKEGHVKMVVELLHKEIILETTTKKGNTALHIAALAGQDEVVRELVNYG
GLNGLHLASKEGHVKMVVELLHKEIILETTTKKGNTALHIAALAGQDEVVRELVNYG


Response time
=============

Requests are handled immediately, in serial order. At present, response
time is quite short and is restricted by the network load rather than CPU
availability. Please note that large output files, such as sometimes occur
with very short query sequences, may need a long time to traverse the net.

Evaluation of the output:
=========================

The output of the server are BLAST search results against the SBASE 
protein domain library. The output file contains the BLAST search
results, organized as follows:

1) List of the best scoring domain entries. As SBASE entries are named 
by domain names (function, structure, etc.), this list already may give 
some information on the expected domain composition.

2) List of alignments. For each SBASE entry you will find the complete 
annotation of the domain, followed by one (or several) alignments with 
(different parts of) the query. If one domain is found several times in 
your query, you may find several alignments with the same or related 
entries at different parts of the query. Please note that it depends on the 
score parameter whether or not you see all the alignments. Do not use very 
low cutoff values because that results in prohibitively long output 
files. (For the time being, we have set the default cutoff to 35 and 
the minimum cutoff to 30).

3) Run statistics. This is usually not essential for the evaluation of 
the results; you can get a complete description of these and other blast 
parameters by sending a HELP message to blast.ncbi.nlm.nih.gov

Important: Failure to see a homology with a known domain may be due to 
several reasons: i) The domain type is not (yet) included in the SBASE 
domain library; ii) The threshold score parameter was set too high for the 
domain to be detected; iii) A different scoring matrix may be necessary in 
order to detect the alignment with the domain type in question. In the present 
experimental version of the SBASE server we support only the matrices used 
by BLAST; "customized matrices" will be added later to the final version.

Papers to reference in reporting results:
=========================================

Pongor, S., Skerl, V., Cserzo, M. and Hatsagi, Z., Simon, G. and 
Bevilacqua, V. (1992): The SBASE domain library release 2.0&  A 
collection of annotated protein sequence segments, Nucleic Acids. Res , 
21, 311-315

Altschul, Stephen F., Warren Gish, Webb Miller, Eugene W. Myers,
and David J. Lipman (1990) Basic local alignment search tool J. Mol. 
Biol. 215:403-410.

Software availability
=====================

The Sbase database is available by anonymous ftp at
ftp.icgeb.trieste.it:/pub/sbase2

The blast software is available by anonymous ftp at
ncbi.nlm.nih.gov:/pub/


Protection of your sequence data
================================

The query sequences are not stored in any form. 

Further info:
=============

Server functions:   Zsolt Hatsagi 
                    <hatsagi at icgeb.trieste.it> 
                    Tel: +39-40-3757342

                    Valeria Bevilacqua, systems manager
                    (valeria at icgeb.trieste.it)
                    Tel.: +39-40-3757330

General info:       Sandor Pongor
                    <pongor at icgeb.trieste.it>
                    Tel: +39-40-3757300

FAX:                +39-40-226-555

Mail:	    	    International Centre for Genetic Engineering
    	    	    and Biotechnology
    	    	    AREA Science Park, Padriciano 99
    	    	    34012 Trieste, Italy



More information about the Plantbio mailing list