[
Worm Breeder's Gazette,
1990]
A number of improvements have been made in the gm automated DNA sequence analysis system: (1) The ratio of AT-containing to CG-containing dinucleotides has been added as a test for introns. This works better than AT frequency alone in C. elegans.(2) A branch site consensus sequence or an enhanced dinucleotide ratio can be required as an additional test on introns. (3) Predicted amino-acid sequence files are generated in a format appropriate for input to the Dana-Farber motif-identification program plsearch. (4) A graphic interface based on X-windows, version 11 is available as an option. (5) The complete analysis algorithm is significantly faster than the previous version. (6) A greedy model evaluation algorithm is available as an option. This algorithm generates the longest, non-overlapping models that cover a sequence and is much faster than the complete analysis algorithm. The program has been tested on Sun3, Sun4 and VAX machines running Unix (Ultrix on the VAX). Results for a series of tests run on a Sun 4/60 are shown in the table. [See Figure 1] gm can be run remotely on our machine, using the Internet. To do this, telnet to haywire.nmsu.edu, and logon as gm_guest with password gmuser. Read the README file for information on running gm. We are also distributing gm as C source code to nonprofit laboratories, either via anonymous ftp to haywire.nmsu.edu, or on tape. If you would like to receive gm on tape, send a 1/4' cartridge or 1/2' reel tape to: Chris Fields, Box 30001/3CRL, New Mexico State University, Las Cruces, New Mexico 88003-0001, USA; Telephone (505) 646-2848.