Eleanor Dodson wrote:
Well - the old way to estimate sigma was Sqrt(I**2 +
constant_representing background) and then get
Sigma_F as ~ sqrt(SigI)/(2*F ) .
sftools would calculate that for you and append it to the output file..
Eleanor
I suggested something similar to Pete but sufficiently different that
I'd like to post it. I expect that in the example above I**2 should be
F**2 otherwise SigI is proportional to I when I >> background rather
than proportional to Sqrt(I) as expected for pure counting statistics.
Bart
For any purist there is no good way. If you look for something that you
can explain concisely in a methods section and that has at least some
logic to it you could convert your Fcalc to intensities (F**2). Multiply
this by a conversion factor C with C being four divided by the average
intensity at the highest resolution of the data set (4/<Ihr>). Take the
square root of this as your SigmaI.
The idea here is to convert the calculated intensity to photon counts
recorded in the experiment. A "typical" data set has about I/SigI of 2
at the high resolution. Since SigI is the square root of I if it is
solely dependent on counting statistics, setting C to give an average I
of 4 in the highest resolution shell should give an I/SigI of about two
after you set SigI to the square root of I.
I don't actually expect that this will closely mimic an experimental
I/SigI versus resolution pattern but it should be easy to calculate with
sftools so you can go ahead and give it a try.
Bart
Eleanor Peter Adrian Meyer wrote:
I add a fake "sigma" column for each "data" column because so many
programs require one.
This is slightly tangential, but does anyone know of a good way to
generate semi-realistic sigma values for calculated/simulated data?
The best I've been able to do is borrow from an experimental dataset of
the same protein (after scaling), but that doesn't work unless you've got
an experimental dataset corresponding to your simulated one. I also
tried
a least-squares fit (following a reference I don't have in front of
me...this was a while ago), which didn't result in a good fit for our
data.
Pete
Pete Meyer
Fu Lab
BMCB grad student
Cornell University
--
==============================================================================
Bart Hazes (Assistant Professor)
Dept. of Medical Microbiology & Immunology
University of Alberta
1-15 Medical Sciences Building
Edmonton, Alberta
Canada, T6G 2H7
phone: 1-780-492-0042
fax: 1-780-492-7521
==============================================================================
