Skip to content

Chemical exchange

Deep-learning CEST analysis

Extract exchange parameters from CEST profiles with a trained network, without hand-tuned starting values.

Conventional CEST fitting needs a starting model and reasonable initial parameters, and getting those wrong wastes time or produces a confident local minimum. A network trained on simulated profiles across the parameter space maps a profile to parameters directly, with no initialisation to choose.

It is fast enough to run across every residue as a first pass, which makes it most useful as a way to find the residues worth fitting properly and to generate the starting values that conventional fitting then needs.

What it needs

  • CEST series
  • Peak list or assignment

What you get

  • Exchange parameters per residue
  • Starting values for a conventional fit

Enter wherever your data already is

This analysis runs as a full pipeline from spectra, or picks up from work you have already done — so nothing has to be repeated to get to the next step.

  • From spectra
  • From a peak list
  • From fitted intensities
Enter a pipeline wherever your data already isA four-stage pipeline runs left to right: spectra, peak picking, peak fitting, rate fitting. Three labelled entry points feed in from below — full starts at spectra, from-picking starts at peak fitting, and from-fitting starts at rate fitting — so work already done elsewhere does not have to be repeated.SpectraPeak pickingPeak fittingRate fittingfullfrom-pickingfrom-fitting

What does the work

  • DNNautoCEST

    Open source, redistributed under its licenceMIT-equivalent: written consent from D. Flemming Hansen

    Trained CEST analysis

Related in Chemical exchange