Towards autonomous analysis of Chemical Exchange Saturation Transfer experiments using Deep Neural Networks

Abstract

Macromolecules often exchange between functional states on timescales that can be accessed with NMR spectroscopy and many NMR tools have been developed to characterise the kinetics and thermodynamics of the exchange processes, as well as the structure of the conformers that are involved. However, analysis of the NMR data that report on exchanging macromolecules often hinges on complex least-squares fitting procedures as well as human experience and intuition, which, in some cases, limits the widespread use of the methods. The applications of deep neural networks (DNNs) and artificial intelligence have increased significantly in the sciences, and recently, specifically, within the field of biomolecular NMR, where DNNs are now available for tasks such as the reconstruction of sparsely sampled spectra, peak picking, and virtual decoupling. Here we present a DNN for the analysis of chemical exchange saturation transfer (CEST) data reporting on two- or three-site chemical exchange involving sparse state lifetimes of between approximately 3 - 60 ms, the range most frequently observed via experiment. The work presented here focuses on the 1H CEST class of methods that are further complicated, in relation to applications to other nuclei, by anti-phase features. The developed DNNs accurately predict the chemical shifts of nuclei in the exchanging species directly from anti-phase 1HN CEST profiles, along with an uncertainty associated with the predictions. The performance of the DNN was quantitatively assessed using both synthetic and experimental anti-phase CEST profiles. The assessments show that the DNN accurately determines chemical shifts and their associated uncertainties. The DNNs developed here do not contain any parameters for the end-user to adjust and the method therefore allows for autonomous analysis of complex NMR data that report on conformational exchange.

Optimizing frequency sampling in CEST experiments

“…In a conventional 1D experiment it is well established that acquisition times should not extend beyond where signal persists. Sampling beyond this limit adds noise , where cest_baseline is the fitted baseline, followed by real Fourier transformation (Karunanithy et al, 2021). This is illustrated schematically in Figure 1a for a simulated profile generated for an exchanging system with parameters as indicated.…”

“…For this sampling interval the time domain signal extends until the end of the acquisition period, as might be the case had the data being optimally recorded in the time domain to start with. Note that a real discrete Fourier transform of a CEST profile generated with frequency sampling of Dvrf = sw/N between each CEST point, where sw and N are the spectral width of the region sampled and the total number of sampled frequency points, respectively, produces a time domain signal with approximately N/2 non-redundant real and imaginary points (N/2+1 and (N+1)/2 for even and odd N, respectively, Figure S1) (Karunanithy et al, 2021). In what follows, we display time domain signals of only the nonredundant points, extending to N/(2sw) (even N) or (N-1)/(2sw) (odd N) in the time-domain.…”