> From MEWES at MIPS17.dnet.mips.biochem.mpg.de Tue May 21 04:41:00 1996
> F. Oulette wrote:
>> >>but I agree with you that it should be somewhere in the genome
>> If the genetic data are correct, the gene should be located on the
> right arm of chr. IV.
>> We knew the fact that ZIP1 was not on the sequence released (Francis,
> it is easy to test this without GenBank annotation!).
yes, this is why I was able to see it was missing from the raw
sequence :)
> One lab is already
> looking into the matter. Due to the described effects of the disruption
> the fact that we don't find ZIP1 or a closely related gene is puzzling.
> Also SUC2 is not found on the genome but this fact can be explained by
> strain variations.
that SUC2, and enzyme involved in sugar metabolism, is missing from an
S288c related strain is not new. That ZIP1, which Ken Wolfe says (as
per the litterature) is involved in "synaptonemal complex protein
required for meiotic chromosome synapsis" is something new, which is
not explained by 3 errors/10 kb!
> The 'updates and corrections to come' are well expected. Detailed
> investigations on CHRXI have shown that the expected error rate is
> around 3 errors/ 10kBases (with a majority of frame-shifts). We can
> expect 3000 errors in the sequence.
We are talking about 3091 bp being missed at a single place in the
yeast genome, not a nucleotide here and there!
> However, if one compares the results
> of the systematic sequencing effort to the earlier published sequences,
> the error rate is significantly lower (strain differences
> may be responsible for some of the differences, but the 'old' sequences
> show at least 2 times more errors). Unfortunatly we have to live with
> unreliable information in the nucleic acid databases. Despite the 99.7%
> of bases that are correct, we cannot distinguish the bad from the good.
I am not challanging the accuracy of the work done done by the genome
sequencing group (I think it is extremely good, I even contributed to
some of it myself!), and I'm actually not challanging anything. I was
just stating that Ken Wolfe's observation was interesting, and that we
will probably see others like it. Mike Cherry correctly pointed out to
me that such observation was only possible because the complete yeast
genome was finished.
> Only independent sequencing will significantly lower the error rate. But
> this is to expensive for the moment. Of course, major mistakes as lost
> fragments, misassembly etc. are different events than individual sequencing
> errors. The piece of DNA we are looking for in the case of ZIP1 is at
> least 3kBases long.
LOCUS YSCZIP1A 3091 bp DNA PLN 26-MAR-1993
DEFINITION Saccharomyces cerevisiae ZIP1 protein gene, complete cds.
ACCESSION L06487
yes, that is what we see from the GenBank record! This could be a
cloning problem, or there could be some other biological explanation.
I'm not saying it one way or the other.
All the best,
francis
--
| B.F. Francis Ouellette | tel: (301) 496-2477 ext 247 |
| GenBank | fax: (301) 435-2433 |
| | NCBI/NLM/NIH Building 38A |
|francis at ncbi.nlm.nih.gov | Bethesda, MD 20894, USA |
