Manuel Cordova^1*, Martins Balodis¹, Albert Hofstetter¹, Federico Paruzzo¹, Sten O. Nilsson Lill², Emma S. E. Eriksson², Pierrick Berruyer¹, Bruno Simões de Almeida¹, Michael J. Quayle³, Stefan T. Norberg⁴, Anna Svensk Ankarberg⁴, Staffan Schantz^4*, Lyndon Emsley^1*

1 Institut des Sciences et Ingénierie Chimiques, École Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland

2 Early Product Development and Manufacturing, Pharmaceutical Sciences, R&D, AstraZeneca, Gothenburg, Sweden

3 New Modalities and Parenteral Development, Pharmaceutical Technology & Development, Operations, AstraZeneca, Gothenburg, Sweden

4 Oral Product Development, Pharmaceutical Technology & Development, Operations, AstraZeneca, Gothenburg, Sweden

* Corresponding authors emails: manuel.cordova@epfl.ch, Staffan.Schantz@astrazeneca.com, lyndon.emsley@epfl.ch

DOI10.24435/materialscloud:gg-mx [version v1]

Publication date: Mar 16, 2021

How to cite this record

Manuel Cordova, Martins Balodis, Albert Hofstetter, Federico Paruzzo, Sten O. Nilsson Lill, Emma S. E. Eriksson, Pierrick Berruyer, Bruno Simões de Almeida, Michael J. Quayle, Stefan T. Norberg, Anna Svensk Ankarberg, Staffan Schantz, Lyndon Emsley, Structure determination of an amorphous drug through large-scale NMR predictions, Materials Cloud Archive 2021.41 (2021), https://doi.org/10.24435/materialscloud:gg-mx

Description

Knowledge of the structure of amorphous solids can direct, for example, the optimization of pharmaceutical formulations, but atomic-level structure determination in amorphous molecular solids has so far not been possible. Solid-state NMR is among the most popular methods to characterize amorphous materials, and Molecular Dynamics (MD) simulations can help describe the structure of disordered materials. However, directly relating MD to NMR experiments in molecular solids has been out of reach until now because of the large size of these simulations. Here, using a machine learning model of chemical shifts, we determine the atomic-level structure of the hydrated amorphous drug AZD5718 by combining dynamic nuclear polarization-enhanced solid-state NMR experiments with predicted chemical shifts for MD simulations of large systems. From these amorphous structures we then identify H-bonding motifs and relate them to local intermolecular complex formation energies.

Materials Cloud sections using this data

Files

File name	Size	Description
README.md MD5md5:6e2e913b9198e679bc7842bcfd13c230	10.2 KiB	Detailed description of the files in the dataset
Scripts.zip MD5md5:460f3702015e81f2d2bde0f88290d369	1.3 MiB	Python notebooks used for data analysis
Figures.zip MD5md5:077feccf2c4063778983f94768beae7a	350.8 KiB	Figures generated by the Python scripts
NMR_Experiments.zip MD5md5:51cf1a14b077222915b89908397d9041	444.9 MiB	Raw NMR data
CIFs.zip MD5md5:057ef7a36ff055313036d94b8a281e34	46.8 KiB	CIF files of candidate crystal structures, XRD structure, potential tautomers, and perturbed crystal structure.
HMBI.zip MD5md5:f68f06304dd00de678cdd4ddf434653b	1.7 MiB	Chemical shift computation of the candidate crystal structures
MD_Snapshots.zip MD5md5:f90aa5f0e1341a01022fc0e9c512dcad	2.7 GiB	Molecular Dynamics trajectories of amorphous AZD5718 with different water contents
ML_Shifts.zip MD5md5:2c18832fa57724b5b9a25dd34ad52632	2.3 GiB	Predicted shieldings for all snapshots of the MD trajectories
DFTB_D3H5.zip MD5md5:44b01089f5d0f7bb01a40910ce800616	1.3 MiB	Formation energies of molecules in the MD trajectories

License

Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.

External references

Keywords

MARVEL/DD1 SNSF machine learning molecular dynamics Experimental NMR