CUPID Paper: Experimental Data & Result Scripts

This repository provides the experimental datasets and results which are presented in the following paper:

Full-Signal Ultrahigh-Resolution NMR by Parameter Estimation

Simon G. Hulse¹ https://orcid.org/0000-0002-5822-6198

Mathias Nilsson² https://orcid.org/0000-0003-3301-7899

Gareth A. Morris² https://orcid.org/0000-0002-4859-6259

Mohammadali Foroozandeh¹ https://orcid.org/0000-0001-8937-3118

¹Chemistry Research Laboratory, University of Oxford, 12 Mansfield Rd, Oxford, OX1 3TA, UK

²Department of Chemistry, University of Manchester, Oxford Road, Manchester M13 9PL, UK

^*mohammadali.foroozandeh@chem.ox.ac.uk

<INSERT CITATION HERE>

Datasets

• data/estradiol/ - 17β-estradiol dataset (main paper Figure 3):

  • 1/ - 2DJ dataset.
  • 2/ - PSYCHE dataset.
  • 3/ - 2DJ dataset acquired using a higher-concentration sample.

• data/quinine/1 - Quinine 2DJ dataset (main paper Figure 2):

  • pdata/1/ - Processed 2DJ spectrum.
  • pdata/2/ - Pure-shift spectrum produced using a 45° tilt and projection.

• data/four-multiplets/ - Simulated datasets presented as part of the ``Four Multiplets'' result in Figure 6 of the Supporting Information (SI):

  • run.py - Python script called to make the data.
  • make_data.m - MATLAB script which makes a single 2DJ dataset with Spinach.
  • pathdef.m - MATLAB script which defines the path. I found that the easiest way to run MATLAB from the command line was to simply place this in the working directory.
  • jres_seq.m - 2DJ Spinach pulse sequence.
  • datasets/dataset_[1-5].mat - The resultant datasets. These can be imported into Python using scipy.io.loadmat.
  • datasets/dataset_info.toml - The shifts and couplings associated with each dataset.

• data/strychnine/ - Simulated strychinine 2DJ dataset (SI Figure 8):

  • make_data.m - MATLAB script which simulates the dataset.
  • run.fish - Fish shell command to run make_data.m from the command line.
  • fid_2dj.mat - 2DJ dataset.
  • pathdef.mat - See above.
  • jres_seq.m - See above.

• data/camphor/1/ - Camphor 2DJ dataset (SI Figure 9):

  • pdata/1/ - Processed 2DJ data.
  • pdata/2/ - Pure-shift spectrum produced using a 45° tilt and projection.

• data/dexamethasone/ - Dexamethasone data (SI Figure 10):

  • 1/ - 2DJ dataset.
  • 2/ - TSE-PSYCHE dataset.

Jupyter Notebooks and Results

The results presented in the paper can be reproduced by running the Jupyter notebooks within the code/ directory.

You are advised to look at the Quinine notebook first, as detailed descriptions are provided at each step of the process.

To run a particular notebook, make sure that you launch Jupyter Lab inside one of the directories listed below. I.e. to look at the quinine notebook, make sure you are in code/quinine/.

• code/quinine/ - Quinine result (main paper Figure 2).

• code/estradiol/ - 17β-Estradiol result (main paper Figure 3).

• code/four-multiplets/ - "Four multiplets" result (SI Figure 6).

• code/strychnine/ - Simulated strychnine result (SI Figure 8).

• code/camphor/ - Camphor result, (SI Figure 9).

• code/dexamethasone/ - Dexamethasone result, (SI Figure 10).

• code/reviewer-1/ - Some miscellaneous notebooks provided in response to comments from Reviewer 1.

  • camphor/ - Considering a region of the camphor dataset.
  • glucose/ - Consideration of a D-(+)-glucose dataset, obtained here
  • estradiol-noise-tolerance - Running CUPID on instances of the estradiol dataset which have been corrupted by synthetic noise.