gwilliam@CRC.AC.UK (Gary Williams x3294) (02/06/90)
Hallo, We are one of the groups using SUN Microsystems (UK)'s implementation of GCG on UNIX machines. We would like to add our own list of requirements, just to keep the momentum going. We really need a decent multiple alignment program, like 'CLUSTAL' or any of the other programs that are springing up around the place, preferably integrated with 'LINEUP'. [Yes, I know you can use 'GAP /OUT' but it gets exceedingly tedious after the n'th sequence!] What about a program for helping to select optimal oligonucleotides for filter hybridisation, sequencing and *in vitro* amplification of DNA, as described in Rychlik W., Rhoads R. (1989) NAR 17, no.21 pp. 8543-8550 What about adding the Accession numbers to the *.SEQCAT one-line descriptions of entries so that when a papers gives a reference to a sequence by its Accession number, 'STRINGS' can be used to quickly find its current entry-name. Indeed, what about being able to reference an entry by its Accession number instead of its entry-name. eg. 'FETCH EMBL:M12345'. [A special plea for UNIX implementations: The SwissProt database has a '$' as part of its entry names. When a '$' occurs as part of a reference to a file in UNIX it means take whatever follows and treat it as a logical symbol to be dereferenced. Why not make things easy for the growing bands of UNIX people and when the 'SwissProtToNBRF' program is written, include a line that changes the '$'s in the entry names to underscores, or whatever, so that when they are 'FETCH'd into a file, the filename means something sensible and doesn't have a '$' in it derived from the entry-name.] OK - SO FLAME ME! Gary Williams Computing Services Section, Janet: G.Williams@UK.AC.CRC MRC-Clinical Research Centre, Elsewhere: G.Williams@CRC.AC.UK Watford Rd, HARROW, Middx, HA1 3UJ, U.K. EARN/Bitnet: G.Williams%CRC@UKACRL Tel 01-869 3294 Fax 01-423 1275 Usenet: ...!mcvax!ukc!mrccrc!G.Williams