Signal sequence prediction

Don Gilbert gilbertd at
Wed Dec 2 17:45:02 EST 1992

In article <1992Dec2.164114.1 at> michael at writes:
>A very long time ago, I got hold of and used a program that could predict
>the occurence of consensus signal sequences in protein sequences (called
>sig_seq.exe ???).

Here is an example of how to search for software when you only know some
key words.  At the IUBio archive, all of the software readme and help
files are index for searching thru gopher.  You remember key words like 
"predict signal sequence".

Here is how the modern biologist can go to his electronic library, and
search out software titles:

Go to the biology software and data library at Indiana U:

      1.  About-IUBio-Gopher  [21Jun92, 3kb].
      2.  About-New-Features  [ 1Nov92, 3kb].
      3.  Drosophila/
      4.  Genbank-Sequences/
 -->  5.  IUBio-Software+Data/

Since you want some software help, look in the software+data section:

 -->  1.  Search IUBio archive ReadMes, Docs & Help files <?>
      2.  Archive.doc  [17Dec91, 16kb].
      3.  File listing  [11Nov92, 71kb].
      4.  New files listing  [11Nov92, 12kb].
      5.  biology/

Since you don't remember the name of the software, try asking the

  Search IUBio archive ReadMes, Docs & Help files: 
      signal and sequence and predict

Here are the results of that search:

      1.  0reviews.txt   /bio/archive/molbio/mac/.
      2.  pm-macinmolbio.txt   /bio/archive/molbio/mac/.
      3.   /bio/archive/help/.
      4.   /bio/archive/help/.
 -->  5.   /bio/archive/help/embl/.
      6.  analyze-signalase.readme   /bio/archive/molbio/mac/.

I happen to know that the first 2 are long reviews of various software.
They are probably worth some of your time to scan thru if you haven't.
Item 3 doesn't sound like what we want.  Item 4 might or might not help.
Item 6 sounds promising, if you have a Macintosh, and item 5 sounds like
help from the EMBL archive.

Well, looking at item 6, we see:

This section is from the document '//IUBio-Software+Data/molbio/mac/analyze-sign

This program is designed for use on the macintosh.  It can be decoded
using binhex or stuffit.  For further information, contact Ned Mantei at
It uses a weight-matrix method to try to predict the site at which signal 
peptides in secretory peptides are cut off by the signal peptidase.

And looking thru item 5, we find this tidbit:

filename             SIGSEQ.UUE
update               11-Dec-1989
version              1.0
size                 60 k
compressed           PAK
ASCII conversion     UUE
author               Popowicz A.M.
description          Prediction of signal sequence cleavage site
literature           CABIOS 4 (1988), 405-406
source code          C
sw requirements      (Compiler for recompiling)
hw requirements      -

The top of item 5 tells us how to get SIGSEQ.UUE if we want:
This section is from the document '//IUBio-Software+Data/help/embl/

More information about the Proteins mailing list