The molecular evolution inference system

 #####   ######  ######  #####   ######   ####   #####   ######   ####    #####
 #    #  #       #       #    #  #       #    #  #    #  #       #          #
 #    #  #####   #####   #    #  #####   #    #  #    #  #####    ####      #
 #    #  #       #       #####   #       #    #  #####   #            #     #
 #    #  #       #       #       #       #    #  #   #   #       #    #     #
 #####   ######  ######  #       #        ####   #    #  ######   ####      #

					### Prototype version (Rev. 0.72) ###
						August 14, 1997

				Copyright (C) 1996-1997 Satoshi OOta

1. What is `DeepForest'?

`DeepForest' is a set of predicates to infer molecular phylogenetic
trees mainly from amino acid sequences. By default, users can use the
maximum likelihood method to infer the trees.  The complete version of
`DeepForest' will analyze sequence data set intelligently, and infers
a reliable molecular phylogenetic tree.

2. Restriction of the present version

This version is a prototype of `DeepForest'. User can use only a part
of function on `DeepForest', and I have to warn you that the present
version is extremely slow. This programs are written almost by KL1,
and can be compiled by KLIC.

3. How to compile `DeepForest'

In directory ./unix, type `test.com'.  All modulues will be generated. 
Of course, you have to install the newest version of klic before the
compilation.

4. How to run demo of prototype of `DeepForest'

A simple demo program `test.kl1' has been prepared. This program tests
almost function of `DeepForest'. You can refer this program to write
your own programs using modules.

The way to run the program is quite simple.  Just type `test'.  The
program will begin to evaluate a given data sets (the data have `.pl'
extensions basically. Pleases refere them to understand the function).

The `test.com' will not compile all source files. `*.com' files are
script files for compiling. Please compile remainders by hand using
the `*.com' files. I shoud note that these programs are experimental.

Binary files (for Soralis 2.3) have been prepared. If you are Sun
SparcStation/Soralis 2.x user, please copy these binary files to an
execution directory. Once test.com is executed, these binary files will
be destroyed.

5. Parallel execution of `DeepForest'

Parallel logic programming made `DeepForest' executable in
distributed/shared memory parallel environment.  If distributed/shared
memory klic is available in you system, why don't you try parallel
execution by `DeepForest'? I have tested parallel system on three
clustered workstations (distributed version) and CRAY CS6400 (shared
memory version).

6. Required environment

This prototype version is so small that you can run `DeepForest' even
on PC with the newest version of klic.  I have tested on Macintosh
(Duo 280c) with Mach Ten.  Also PIMOS version of `go/0' (contour/1 and
traverse/3) is included in this file.  Since these programs are quite
experimental one, only nucleotide data are available.

7. Concept of `DeepForest'

`DeepForest' is not only  an  application   program,   but   also
set   of  predicates   which   are  useful for data analyses. The
programs  are  presented  as  modules,  which   includes   useful
predicates,  and  users  can   construct  new customized programs
easily in scheme of logic programming.

8. Documentation

For more information, please refer the documents under
./df/doc/DeepForest directory.

------------------------------------------------------------
Satoshi OOta (Neither Ota nor Ohta nor Oota, but OOta!)
------------------------------------------------------------
Department of Genetics,
School of Life Science,
The Graduate University of Advanced Studies
at
Laboratory of Evolutionary Genetics,
National Institute of Genetics,
Mishima, 411 Japan
FAX: +81-559-81-6794
TEL: +81-559-81-6790
------------------------------------------------------------

				Copyright (C) 1996-1997 Satoshi OOta

