The Library
Combining machine learning and molecular simulations to predict the stability of amorphous drugs
Tools
Sosso, Gabriele C. and Barnard, Trent (2023) Combining machine learning and molecular simulations to predict the stability of amorphous drugs. Journal of Chemical Physics, 159 (1). 014503. doi:10.1063/5.0156222 ISSN 0021-9606.
|
PDF
014503_1_5.0156222.pdf - Published Version - Requires a PDF viewer. Available under License Creative Commons Attribution 4.0. Download (8Mb) | Preview |
|
PDF
Accepted_Manuscript.pdf - Accepted Version Embargoed item. Restricted access to Repository staff only - Requires a PDF viewer. Download (2423Kb) |
Official URL: https://doi.org/10.1063/5.0156222
Abstract
Amorphous drugs represent an intriguing option to bypass the low solubility of many crystalline formulations of phar- maceuticals. The physical stability of the amorphous phase with respect to the crystal is crucial to bring amorphous formulations into the market - however, predicting the timescale involved with the onset of crystallisation a priori is a formidably challenging task. Machine learning can help in this context, by crafting models capable of predicting the physical stability of any given amorphous drug. In this work, we leverage the outcomes of molecular dynamics sim- ulations to further the state-of-the-art. In particular, we devise, compute and use ”solid state” descriptors that capture the dynamical properties of the amorphous phases, thus complementing the picture offered by the ”traditional”, ”one- molecule” descriptors used in most quantitative structure–activity relationship models (QSAR) models. The results in terms of accuracy are very encouraging, and demonstrate the added value of using molecular simulations as a tool to
enrich the traditional machine learning paradigm for
I. INTRODUCTION
Most modern pharmaceutical drugs are packaged as crys- talline formulations1. The crystalline structure has signifi- cant effects on several physical properties of the drug, such as its solubility, its stability and its bioavailability2. Cru- cially, almost 90% of pharmaceutical drugs are categorised as poorly water soluble3,4, which clearly limits their effective- ness, chiefly in terms of bioavailability.
Packaging pharmaceutical drugs as amorphous formula- tions represents a viable way forward in order to improve the solubility of modern drug formulations5, as they present sev- eral benefits in comparison to crystalline drugs. Firstly, most amorphous compounds are intrinsically much more soluble than their crystalline counterparts6–8. As such, amorphous drugs typically act more quickly than crystalline drugs9,10. In addition, amorphous drugs can be more easily packaged into different formulations - such as tablets, capsules, or suspen- sions8,11. In fact, the lack of crystalline structure can also al- low for greater flexibility in designing drug delivery systems with specific properties, such as sustained release or targeted delivery8 .
While amorphous drugs appear to have an edge over their crystalline counterparts, they also have some disadvantages that can make their development and formulation challenging - chiefly their lack of stability. Amorphous solids are almost always metastable with respect to their crystalline phases, which means that amorphous drugs have a tendency to crys- tallise12 - within a timescale that is very challenging to predict. This represents a serious problem12, in that the properties of the crystalline form might differ from that of the amorphous phase - which poses a severe clinical risk. In addition, the structural relaxation of the glass alone might alter the func- tional properties of the amorphous formulation13. It is also important to note that the production of amorphous drugs can
a)Corresponding author: g.sosso@warwick.ac.uk
drug design and discovery.
Item Type: | Journal Article | ||||||
---|---|---|---|---|---|---|---|
Subjects: | Q Science > QC Physics Q Science > QD Chemistry |
||||||
Divisions: | Faculty of Science, Engineering and Medicine > Science > Chemistry | ||||||
Library of Congress Subject Headings (LCSH): | Drugs -- Solubility, Solubility, Machine learning, Pharmaceutical technology, Crystallization | ||||||
Journal or Publication Title: | Journal of Chemical Physics | ||||||
Publisher: | American Institute of Physics | ||||||
ISSN: | 0021-9606 | ||||||
Official Date: | 3 July 2023 | ||||||
Dates: |
|
||||||
Volume: | 159 | ||||||
Number: | 1 | ||||||
Article Number: | 014503 | ||||||
DOI: | 10.1063/5.0156222 | ||||||
Status: | Peer Reviewed | ||||||
Publication Status: | Published | ||||||
Access rights to Published version: | Open Access (Creative Commons) | ||||||
Date of first compliant deposit: | 11 July 2023 | ||||||
Date of first compliant Open Access: | 11 July 2023 | ||||||
RIOXX Funder/Project Grant: |
|
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |
Downloads
Downloads per month over past year