Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study
In response to the increasing application of machine learning (ML) across many facets of pharmaceutical development, this pilot study investigated if ML, using artificial neural networks (ANNs), could predict the apparent degree of supersaturation (aDS) from two supersaturated LBFs (sLBFs). Accuracy...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Pharmaceutics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4923/13/9/1398 |
| _version_ | 1797517569425408000 |
|---|---|
| author | Harriet Bennett-Lenane Joseph P. O’Shea Jack D. Murray Alexandra-Roxana Ilie René Holm Martin Kuentz Brendan T. Griffin |
| author_facet | Harriet Bennett-Lenane Joseph P. O’Shea Jack D. Murray Alexandra-Roxana Ilie René Holm Martin Kuentz Brendan T. Griffin |
| author_sort | Harriet Bennett-Lenane |
| collection | DOAJ |
| description | In response to the increasing application of machine learning (ML) across many facets of pharmaceutical development, this pilot study investigated if ML, using artificial neural networks (ANNs), could predict the apparent degree of supersaturation (aDS) from two supersaturated LBFs (sLBFs). Accuracy was compared to partial least squares (PLS) regression models. Equilibrium solubility in Capmul MCM and Maisine CC was obtained for 21 poorly water-soluble drugs at ambient temperature and 60 °C to calculate the aDS ratio. These aDS ratios and drug descriptors were used to train the ML models. When compared, the ANNs outperformed PLS for both sLBF<sub>Capmul</sub><sup>MC</sup> (<i>r</i><sup>2</sup> 0.90 vs. 0.56) and sLBF<sub>Maisine</sub><sup>LC</sup> (<i>r</i><sup>2</sup> 0.83 vs. 0.62), displaying smaller root mean square errors (RMSEs) and residuals upon training and testing. Across all the models, the descriptors involving reactivity and electron density were most important for prediction. This pilot study showed that ML can be employed to predict the propensity for supersaturation in LBFs, but even larger datasets need to be evaluated to draw final conclusions. |
| first_indexed | 2024-03-10T07:18:23Z |
| format | Article |
| id | doaj.art-7b06bc13a1894a9ba8b0eaab5f9147a4 |
| institution | Directory Open Access Journal |
| issn | 1999-4923 |
| language | English |
| last_indexed | 2024-03-10T07:18:23Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Pharmaceutics |
| spelling | doaj.art-7b06bc13a1894a9ba8b0eaab5f9147a42023-11-22T14:47:15ZengMDPI AGPharmaceutics1999-49232021-09-01139139810.3390/pharmaceutics13091398Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot StudyHarriet Bennett-Lenane0Joseph P. O’Shea1Jack D. Murray2Alexandra-Roxana Ilie3René Holm4Martin Kuentz5Brendan T. Griffin6School of Pharmacy, University College Cork, T12 YT20 Cork, IrelandSchool of Pharmacy, University College Cork, T12 YT20 Cork, IrelandSchool of Pharmacy, University College Cork, T12 YT20 Cork, IrelandSchool of Pharmacy, University College Cork, T12 YT20 Cork, IrelandDrug Product Development, Janssen Research and Development, Johnson & Johnson, Turnhoutseweg 30, 2340 Beerse, BelgiumSchool of Life Sciences, University of Applied Sciences and Arts Northwestern Switzerland, Hofackerstrasse 30, 4132 Muttenz, SwitzerlandSchool of Pharmacy, University College Cork, T12 YT20 Cork, IrelandIn response to the increasing application of machine learning (ML) across many facets of pharmaceutical development, this pilot study investigated if ML, using artificial neural networks (ANNs), could predict the apparent degree of supersaturation (aDS) from two supersaturated LBFs (sLBFs). Accuracy was compared to partial least squares (PLS) regression models. Equilibrium solubility in Capmul MCM and Maisine CC was obtained for 21 poorly water-soluble drugs at ambient temperature and 60 °C to calculate the aDS ratio. These aDS ratios and drug descriptors were used to train the ML models. When compared, the ANNs outperformed PLS for both sLBF<sub>Capmul</sub><sup>MC</sup> (<i>r</i><sup>2</sup> 0.90 vs. 0.56) and sLBF<sub>Maisine</sub><sup>LC</sup> (<i>r</i><sup>2</sup> 0.83 vs. 0.62), displaying smaller root mean square errors (RMSEs) and residuals upon training and testing. Across all the models, the descriptors involving reactivity and electron density were most important for prediction. This pilot study showed that ML can be employed to predict the propensity for supersaturation in LBFs, but even larger datasets need to be evaluated to draw final conclusions.https://www.mdpi.com/1999-4923/13/9/1398lipid-based drug deliverycomputational pharmaceuticsmachine learningsupersaturated lipid-based formulations |
| spellingShingle | Harriet Bennett-Lenane Joseph P. O’Shea Jack D. Murray Alexandra-Roxana Ilie René Holm Martin Kuentz Brendan T. Griffin Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study Pharmaceutics lipid-based drug delivery computational pharmaceutics machine learning supersaturated lipid-based formulations |
| title | Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study |
| title_full | Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study |
| title_fullStr | Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study |
| title_full_unstemmed | Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study |
| title_short | Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study |
| title_sort | artificial neural networks to predict the apparent degree of supersaturation in supersaturated lipid based formulations a pilot study |
| topic | lipid-based drug delivery computational pharmaceutics machine learning supersaturated lipid-based formulations |
| url | https://www.mdpi.com/1999-4923/13/9/1398 |
| work_keys_str_mv | AT harrietbennettlenane artificialneuralnetworkstopredicttheapparentdegreeofsupersaturationinsupersaturatedlipidbasedformulationsapilotstudy AT josephposhea artificialneuralnetworkstopredicttheapparentdegreeofsupersaturationinsupersaturatedlipidbasedformulationsapilotstudy AT jackdmurray artificialneuralnetworkstopredicttheapparentdegreeofsupersaturationinsupersaturatedlipidbasedformulationsapilotstudy AT alexandraroxanailie artificialneuralnetworkstopredicttheapparentdegreeofsupersaturationinsupersaturatedlipidbasedformulationsapilotstudy AT reneholm artificialneuralnetworkstopredicttheapparentdegreeofsupersaturationinsupersaturatedlipidbasedformulationsapilotstudy AT martinkuentz artificialneuralnetworkstopredicttheapparentdegreeofsupersaturationinsupersaturatedlipidbasedformulationsapilotstudy AT brendantgriffin artificialneuralnetworkstopredicttheapparentdegreeofsupersaturationinsupersaturatedlipidbasedformulationsapilotstudy |