2 $Name: fa_34_26_5 $ - $Id: readme.mpi_3.3,v 1.4 2001/08/20 21:18:47 wrp Exp $
6 This file is obsolete - see readme.v34t0, readme.v33t0, and
7 readme.pvm_3.4 for more up-to-date information. With version 3.4, the
8 MPI programs are mp34comp*, mu34comp*, etc.
14 This distribution includes the first full-function MPI implementation of
15 the libary-vs-library comparison programs. The following programs are
18 Programs to produce conventional scores and alignments:
20 mp3compfa protein vs protein, DNA vs DNA
21 mp3compsw protein vs protein, DNA vs DNA
22 mp3compfx/ DNA vs protein
23 mp3comptfx/y protein vs DNA
25 Programs to summarize the effectiveness of a search (require
26 super-family-labeled databases):
28 ms3compfa protein vs protein, DNA vs DNA
29 ms3compsw protein vs protein, DNA vs DNA
30 ms3compfx/ DNA vs protein
31 ms3comptfx/y protein vs DNA
33 Programs to report the scores and alignments of the highest scoring
34 unrelated sequence (require super-family-labeled databases). These
35 programs are used to evaluate the super-family labeling.
37 mu3compfa protein vs protein, DNA vs DNA
38 mu3compsw protein vs protein, DNA vs DNA
39 mucompfx/ DNA vs protein
40 mu3comptfx/y protein vs DNA
42 Note that the current parallel implementations distribute the second
43 database among 'N' parallel workers by approximately dividing the
44 database into 'N' parts by seeking into the middle of the database and
45 finding the next entry. This strategy fails when the database is a
46 single long sequence (the first worker gets the entire database, the
49 This version has been tested using the MPICH implementation of MPI,
50 which is available from:
52 ftp://ftp.mcs.anl.gov/mpi
54 See readme.pvm_3.3 for other information about the development of
55 these programs. Both the PVM (pv3compfa, etc.) and MPI (mp3compfa,
56 etc.) sets of programs use the same sets of source files; differences
57 in the two implementations are specified with #define PVM_SRC and