In article <30E560BC.41C67EA6 at bioch.ox.ac.uk> "Simon M. Brocklehurst" <smb at bioch.ox.ac.uk> writes:
>> To be fair to the authors of the program (althought they certainly don't
> need me to defend them), they didn't contribute to all the hype about how
> great LINUS is on bionet. And the fact that more general media got ahold
> of the "solution to the folding problem" story shouldn't carry any weight
> in peoples' minds.
> I don't think Srinivasn and Rose need to be too concerned about how the
> program performs on a wide variety of systems because this is not what they
> published. They (somewhat unusually for computational biology papers) presented
> the results they got in an honest pictorial form in their Proteins paper from
> which people should make their own judgements about the capabilities of this
> particular approach.
> When others get their hands on the program, I hope no-one is expecting to get
> any mileage from saying, "Heh, I thought LINUS was supposed to be a
> ground-breaking piece of software that can really fold proteins up into roughly
> native structures. But we find that it doesn't do that well for a lot of
> proteins" etc etc... The interesting finding, of course, would be that it
> DOES in fact work well. It's hardly a surprise if it doesn't work properly
> is it?
>
I do also like the paper, it was better than most papers. However I
did not like the hype that was created before the paper was
published. I would also be surprised if the authors atleast could have
helped to decrease the amount of hype if they wanted to.
Does anyone know if they are going to distribute source code or only
executables ? In any way I'd be very interested in testing it.
arne
(To think before you speak is as to wipe your ass before you defecate.)
-----------------------------------------------------------------
From: Arne Elofsson Email: arne at biokemi.su.se
Tel:+46(0)8-161553 WWW: http://rune.biokemi.su.se/~arne/
