# DP5 without DFT: uncertainty-calibrated graph neural net accelerates structure confirmation via NMR

## TL;DR

This paper introduces DP5q, a fast machine learning method for assigning molecular structures to NMR spectra, replacing slower DFT calculations.

## Contribution

DP5q uses a graph neural network and quantile regression to accelerate NMR structure confirmation without DFT.

## Key findings

- DP5q achieves comparable accuracy to DFT-based DP5 but is significantly faster.
- The method was validated on thousands of molecules and challenging cases.
- Quantile regression helps calibrate uncertainty in predictions.

## Abstract

The evaluation and assignment of candidate structures to NMR spectra can be facilitated by the DP4 method, which assumes that one of the candidate structures is correct, and the DP5 method, which calculates the probability of a correct assignment for each candidate individually. Both of these methods require DFT calculations and thus a significant amount of computer resources. In this paper we present DP5q, a new version of DP5, which uses a graph convolutional neural network and quantile regression to replace the DFT-based algorithm. This dramatically increases the speed of the calculation at the cost of a modest decrease in accuracy. We demonstrate the efficacy of this rapid calculation both on a test set of thousands of molecules and also on cases selected for the difficulty of assigning the structure.

DP5q calculates the probability that a molecule is correctly assigned to an NMR spectrum. Extensive testing shows the results are comparable in accuracy to DFT-based DP5 and dramatically faster.