tacg: a program for the restriction enzyme analysis of DNA, V 1.7
      by Harry Mangalam, UC Irvine (mangalam@uci.edu, 714 824 4824)

This file announces the availability and describes the basic functionality 
of 'tacg', a command line tool for the restriction enzyme analysis of DNA 
for unix-like operating systems.  

There is a Web page describing it more fully at:
http://hornet.bio.uci.edu/~hjm/projects/tacg/tacg.main.html

There is also a WWW interface to it so that you can try it out and see if
it's worth downloading:
http://hornet.bio.uci.edu/~hjm/projects/tacg/tacg.form.html

The complete code to implement the Web interface is also available (tho not as 
well debugged as the program itself) at:
http://hornet.bio.uci.edu/~hjm/projects/tacg/tacgi

You can always get the latest version in ANSI C source code and several
pre-compiled binaries (unless I move) at:
ftp://mamba.bio.uci.edu/pub/tacg/


Binaries currently exist for (and appear to work well for):
- IRIX (5.3)
- SunOS (5.3, 4.1.3)
- DEC Unix (aka OSF/1) (V3.0/347)
- Linux(1.2.8, 2.0.0)

It also has been compiled on ConvexOS, DEC Ultrix(MIPS), and NeXTSTEP 
(black), but since these binaries were not very popular with the previous 
releases, I've not pkged them again.  If you want a binary release of one 
of these, give me a shout and I'll make one up for you.

Others will be made available as I find systems on which to compile them, or
as others contribute binaries.  The source code is available in a separate 
distribution; it is made available for examination and porting purposes,  but
is still very crude.


INSTALLATION:
If you've obtained this file separately from the distribution pkg 
(a gzipped tar file, called "tacg-<version>-<hardware/OS>.tar.gz"), 
please move the distribution to a writable-by-you directory, and unpack it:

    gunzip tacg-<version>-<hardware.OS>.tar.gz
      and then
    tar -tvf tacg-<version>-<hardware.OS>.tar (to see what it contains)
      or
    tar -xvf tacg-<version>-<hardware.OS>.tar (to unpack it)

====

It will create it's own subdirectory "tacg-<version>" containing these files:

    README.tacg     this file; intro, how to install it
    tacg            required file - executable; CAN be renamed (not included in
                      source code distribution)
and these directories:
    
    DATA/ (all these files should go into your TACGLIB, usually 
            /usr/local/lib/tacg)
      rebase.data     required file - RE database; CANNOT be renamed
      codon.prefs     required file - codon preferences; CANNOT be renamed
      
      Alternative REBASE files, including:
      Amersham.RB    available from Amersham
      Cheap.RB       more than 40 Units/$ (thanks Dana!)
      NEB.RB         available from New England Biolabs
      Promega.RB     available from Promega
      Stratagene.RB  available from Stratagene
      TFD.RB         a compact version of the Transcription Factor Database
      
      
    DOCS 
      tacg.main.html what it sounds like - an HTML-ized manual
      tacg.man.html  the HTML-ized man page
      tacg.1          crude, preformatted man page (put in local man pages)
      
=====
        
Put the components in the right places.

   - Put the binary 'tacg' in with your local binaries.  
      For the system (you must be root), typically in '/usr/local/bin' 
      and make sure that they're executable:
      `chmod +x /where/you/put/it/tacg`

   - Put the data files (rebase.data, codon.prefs, *.RB) somewhere usable:
      either:
         (as root) in a tacg library directory (perhaps /usr/local/lib/tacg) 
        and set the environment variable TACGLIB to point to them ie:
   
         setenv TACGLIB /usr/local/lib/tacg  (for csh, tcsh, etc)
         TACGLIB=/usr/local/lib/tacg         (for sh, bash, etc)
      
      or:
         make copies of them and keep them in your home directory, in a 
         'private' TACGLIB directory, or in the same directory where you keep 
         all your sequences.  tacg will search for these files in the order: 
         current directory ($PWD), home directory ($HOME), TACGLIB directory 
         $TACGLIB).

   - Put the man page (tacg.1) in your local man (1) directory (often 
   /usr/local/man/man1 or /usr/catman/local/man1 or something similar).
   Put the html pages where you put the system html pages (no std place 
   yet, although I put mine in /usr/local/man/html)
   
   That's it.  You may have to open a new window or telnet session so that 
   your shell picks up the existance of tacg (or type `rehash` in tcsh to 
   re-index).

   - type:
   `tacg -h`
   to get a minimal description of how to use the program.  If you need more 
   help, read the file 'tacg.main.html', using your favorite Web browser 
   (verified with Netscape).
  
   - please feel free to email me (mangalam@uci.edu with any questions, 
   criticisms, problems, etc.  If you don't kvetch, it probably won't get 
   fixed.

====
         
USING tacg:

For the impatient: Here's an example of how to use it:

`tacg -h` will always dump a short description of usage.

`tail +44 seq.file | tacg -n6 -o5 -F2 -g100 -c -l -L -w90 >seq.file.map`

Translation: chop off the top 44 lines of 'seq.file' and pipe the resulting 
sequence to tacg, returning info on all 6+ cutters (-n6) that generate 5' 
overlaps (-o5), giving the doubly sorted (-c) fragment sizes (-F2), a 
ladder map (-l) and a gel output (-g100) of those enz's that match, along 
with the default linear restriction map w/ 1 letter, 1 frame translation 
(-L) and write the output 90 characters wide (-w90) to a file called 
'seq.file.map'.

If you just want to view it, pipe it directly into a pager such as 'more' or 
'less' (latter is far better) ie:

tacg -n6 -o5 -F2 -g100 -c -l -L -w90 <seq.file | less

same options as above, but digesting the whole of 'seq.file' and piping the 
output directly into 'less'.  tacg's output has different sections prefixed 
with "==", so you can skip to them by searching for those characters (/==, 
in 'less').


The source code is made freely available for examination and nonprofit 
purposes, although since it is presently in beta release, I would warn 
those who want to try to use it that the source is still very ugly.  I'd like a 
chance to change it based on responses, document it more extensively and 
neaten it up before people start to play with it extensively.  There are 
also some known bugs in it that make it unattractive for incorporation into 
re-entrant code.

The documentation has recently been converted to HTML and a copy of that file
is included in this distribution (tacg.main.html).  Point your favorite 
Web Browser at it.
The most recent information about tacg is, of course, available online:
http://hornet.bio.uci.edu/~hjm/projects/tacg/tacg.main.html

Hope it's of some use.

Cheers
harry