News‎ > ‎

New paper accepted in Proteins: Structure, Function and Bioinformatics

posted Aug 6, 2011, 6:25 PM by Vincent Voelz   [ updated Sep 9, 2011, 7:58 AM by Vincent Voelz ]
Our manuscript "Calculation of Rate Spectra from Noisy Time Series Data" has just been accepted in the journal Proteins (preprint here). In this work we present a new method for calculating distributions of relaxation timescales from kinetic traces.  This advance is especially important to the protein folding community, as interpreting protein folding kinetics has traditionally required fitting to a predetermined mechanism.  Our approach avoids such model bias by computing relaxation timescales directly from the data, a la an inverse Laplace transform.  On the surface, this problem would seem numerically ill-conditioned, but we show how applying various regularization methods can extract relevant timescales extremely accuracy.  Our hope is that this tool be used as a starting point for further testing of mechanistic models.  

Our python code for computing rate spectra is freely available here.
Comments