ShortDistances

A LabVIEW program to calculate distance distributions

from dipolar broadedened EPR spectra

DEER has become very popular to determine interspin distances of 20-50 Å, but is not suitable for much closer distances. Below 20 Å, the dipolar broadening is sufficient to broaden the CW EPR spectrum. In 2001, we showed that the deconvolution of the dipolar broadening can be used to determine interspin distance distributions in double spin-labeled proteins at room temperature as long as the interpsin vector varies slowly (C. Altenbach, K. -J. Oh, R. Trabanino, K. Hideg and W. Hubbell. Estimation of inter-residue distances in spin labeled proteins at physiological temperatures: experimental strategies and practical limitations.Biochemistry. 2001 40:15471-15482). However, the deconvolution analysis has been a complicated process in the past, requiering significant experience and operator intervention. Here we present an new set of software tools to facilitate and automate the determination of distance distributions from dipolar broadening using Tikhonov regularization and nonlinear optimization. (Abstract from biophysical Society Poster, Long Beach, 2008: 2008 Biophysical Society Meeting in Long Beach, CA. Christian Altenbach, Wayne L. Hubbell. Improved Distance Determination from Dipolar Broadening of EPR Spectra, Biophysical J, 94:827, 2008, abstract supplement)

Features

  • Tikhonov regularization with interactively adjustable regularization parameter.
  • Fitting to an analytical distance distribution of up to four components, each with distance, width, and spin percentage as adjustable parameter for each.
  • Additional fitting parameters are the alignment and percent of nonineracting component.
  • Each parameter can be selected for fitting or held constant.
  • Full printed report of asymtotic standard errors in the fitting paramters as well as the correlation matrix.
  • All parameters can be manually adjusted for simulation of the corresponding interacting spectrum.
  • The data used in computations can be a linear combination of the raw data. This allows interactive removal of noninteracting components in (D) due to underlabeling and removal of interacting component is (S) due to the statistics of dilute labeling.

Requirements

Two EPR spectra as single column ASCII files with 2048 points.

(D) The dipolar broadened spectrum. This is typically a spin labeled protein monomer containing two labels, a dimer of a protein where each subunit contains one label, or similar.

(S) The same protein, but without dipolar interaction. This is most easily achieved by spin labeling the protein with a mixture of spin label and a diamagnetic analogue.

* The baseline must be clean (very important).

* There should be no contaminant of free spin label. If there is, it should be removed using one of my other programs.

(The spectra don't need to be normalized for equal integrated area and they don't need to be perfectly aligned. The program will take care of these issues automatically)