wrp@biochsn.acc.Virginia.EDU (William R. Pearson) (01/24/91)
A minor bug in tfasta has been discovered (that hid for about four years). It causes the program to core-dump when it encounters a DNA sequence shorter than 5 nt. (Apparently there are several very short sequences in recent releases from genbank.) The bug affects both the older fasta14 and the newer fasta15, and revised versions of both have been posted for anonymous ftp in uvaarpa.virginia.EDU:/public_access/fasta14d.shar(.Z) or fasta15b.shar(.Z). (The IP address of uvaarpa is 128.143.2.7) Bill Pearson
roy@alanine.phri.nyu.edu (Roy Smith) (01/24/91)
wrp@biochsn.acc.Virginia.EDU (William R. Pearson) writes:
-> [a bug] causes the program to core-dump when it encounters a DNA sequence
-> shorter than 5 nt. (Apparently there are several very short sequences in
-> recent releases from genbank.)
Wait a minute. Are you trying to tell me that there are loci in
GenBank that are 4 nt or shorter?
--
Roy Smith, Public Health Research Institute
455 First Avenue, New York, NY 10016
roy@alanine.phri.nyu.edu -OR- {att,cmcl2,rutgers,hombre}!phri!roy
"Arcane? Did you say arcane? It wouldn't be Unix if it wasn't arcane!"JREES@vax.oxford.ac.uk (01/24/91)
>wrp@biochsn.acc.Virginia.EDU (William R. Pearson) writes: >-> [a bug] causes the program to core-dump when it encounters a DNA sequence >-> shorter than 5 nt. (Apparently there are several very short sequences in >-> recent releases from genbank.) > > Wait a minute. Are you trying to tell me that there are loci in >GenBank that are 4 nt or shorter? >-- >Roy Smith, Public Health Research Institute >455 First Avenue, New York, NY 10016 >roy@alanine.phri.nyu.edu -OR- $att,cmcl2,rutgers,hombre!phri!roy >"Arcane? Did you say arcane? It wouldn't be Unix if it wasn't arcane!" Whether or not there are short sequences I cannot tell, but the bug was certainly there, as shown by an (accidentaly) zero length sequence in EMBL one week last year (only one week) which crashed the GCG version - this was fixed by GCG, I guess the info on the bug did not filter back, appologies Bill, I should have alerted you also. Jasper Rees JRees@Vax.Oxford.ac.uk Seqtest@Wccf.Mit.Edu
rak@sun1.ruf.uni-freiburg.de (Bodo Rak) (01/25/91)
wrp@biochsn.acc.Virginia.EDU (William R. Pearson) writes: >A minor bug in tfasta has been discovered (that hid for about >four years). It causes the program to core-dump when it encounters >a DNA sequence shorter than 5 nt. (Apparently there are several >very short sequences in recent releases from genbank.) >The bug affects both the older fasta14 and the newer fasta15, >and revised versions of both have been posted for anonymous ftp in >uvaarpa.virginia.EDU:/public_access/fasta14d.shar(.Z) or fasta15b.shar(.Z). >(The IP address of uvaarpa is 128.143.2.7) >Bill Pearson We run TFASTA (version 1.4a of February 1990) on a PC and found that the program crashes when it encounters entry ECG24 (AC: K01130) in the EMBL data base. The sequence length of this entry is 5 nt. We had to regularly delete this entry in new releases of the data base. I therefore asked Rainer Fuchs at EMBL a while ago whether it would be possible to remove this entry, which is probably of no informational value to anyone. (BTW: We also use Fasta - in batch runs - on a PC to search the whole data base and find it very convenient. The average run takes about 20 minutes so that quite a number of jobs can be done over night. We mainly use mail-fasta to run our sequences against new entries.) Bodo Rak