This program is designed for use on the macintosh.  It can be decoded
using binhex or stuffit.  For further information, contact Ned Mantei at
BIOCHEMI.CZHETH5A.bitnet.  
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.





AnalyzeSignalase is a Macintosh program for applying the 
algorithm of von Heijne (Nucleic Acids Res, 14:4683-4690, 1986) 
to the prediction and analysis of mammalian signal sequences.  
This program is freeware.
©1988--1992, by
Ned Mantei
Department of Biochemistry
Swiss Federal Institute of Technology
ETH-Zentrum
CH-8092 Zurich, Switzerland
Fax: (0041*)-1-252-8744
E-Mail¤: Biochemi@CZHETH5A
*international code from the U.S.A.
¤Fax preferred initially--I don't often check my E-mail.

A window of 15 amino acids at a time, from -13 to +2 around a 
potential signal peptidase cleavage site, is evaluated to give a 
cleavage "score".  The site, not too far from the N-terminus, with 
the maximal score, is likely to be the site at which the signal 
peptide is split off, assuming the signal sequence is split at all.  
Please note that this short explanation is no substitute for reading 
the original paper!
The program uses one (non-Macintosh type) menu:
A)	Choose a data file.  You can choose a new data file at any 
time.  Prepare the sequence data file(s) as follows:  The 
sequence must be in 1-letter amino acid code.  Remove all 
comments and headers (spaces, dots, and numbers are 
ignored).  Save the file in a "text only" format.  If it comes from 
a larger computer through a modem, it is probably already 
text only.  If you use something like MS Word, choose Save 
As, in the file-saving dialog box click on "File Format...", and 
in the subdialog choose "Text only".  Included with this 
package is a sample data file, sample.pep (it is the sequence 
of rabbit lactase-phlorizin hydrolase).
B)	Analyze a range of 21 sites.  Lists the cleavage site scores for 
21 sites in a row, starting at a site you specify.  For example, if 
you want to look at the scores for every potential site near the 
N-terminus, start with 13 (the closest possible site to the N-
terminus).
C)	Analyze one site.  Each of the 15 amino acids in a particular 
window is assigned a number indicative of how favorable 
that particular amino acid is at that particular position relative 
to the cleavage site.  This option shows the values for the 
amino acids around one particular site you choose (the sum 
of these values gives the score for that site).
D)	Analyze one site and propose mutations.  Similar to (C), but 
shows the two mutations which would maximally raise the 
score for this cleavage site.
E)	Find the site with a maximal score.  Scans through the whole 
sequence and finds the site with the highest score.  Pitfall:  If, 
for example, the highest score (e.g., 8) is for cleavage after 
amino acid 20, but there is also a reasonable score (>6, say) 
for cleavage after amino acid 18, the algorithm cannot be 
relied on to distinguish between these two possibilities.
F)	Find sites with a score ³ a cutoff you select.  Helps to avoid 
the problem indicated in (E).  Choose 4 as the cutoff to start 
with, and you will see all sites with a score ³ 4.
G)	Quit.
H)	Read instructions.  Look again at the short instructions which 
appear when starting the program.
I)	Echo everything to a log file.  You are asked to name a log 
file.  All the text which appears on the screen will also be 
written to this file.