Novel enhanced applications of QSPR models: Temperature dependence of aqueous solubility

Klimenko, Kyrylo; Кузьмин, В. Е.; Ognichenko, L. N.; Gorb, Leonid; Shukla, Manoj K.; Vinas, Natalia; Perkins, Edward J.; Polishchuk, Pavel; Artemenko, Anatoly G.; Leszczyński, Jerzy

doi:10.1002/jcc.24424

Cited by 16 publications

(36 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…. [21] The latter can be exceptionally difficult, as shown by Oprisiu et al 2012Oprisiu et al , 2013 in their attempt to model properties of mixtures and facing challenges to select a robust method for test set selection. Solov'ev et al 2011 researchers tried to overcome the "many-to-many" relationship issue in test set selection that arises from modeling properties of mixtures by doing external 5-fold cross-validation.…”

Section: Introductionmentioning

confidence: 99%

QSPR Modeling of Liquid‐liquid Equilibria in Two‐phase Systems of Water and Ionic Liquid

Klimenko

Inês

Esperança

et al. 2020

Molecular Informatics

View full text Add to dashboard Cite

The increasing application of new ionic liquids (IL) creates the need of liquid‐liquid equilibria data for both miscible and quasi‐immiscible systems. In this study, equilibrium concentrations at different temperatures for ionic liquid+water two‐phase systems were modeled using a Quantitative‐Structure‐Property Relationship (QSPR) method. Data on equilibrium concentrations were taken from the ILThermo Ionic Liquids database, curated and used to make models that predict the weight fraction of water in ionic liquid rich phase and ionic liquid in the aqueous phase as two separate properties. The major modeling challenge stems from the fact that each single IL is characterized by several data points, since equilibrium concentrations are temperature dependent. Thus, new approaches for the detection of potential data point outliers, testing set selection, and quality prediction have been developed. Training set comprised equilibrium concentration data for 67 and 68 ILs in case of water in IL and IL in water modeling, respectively. SiRMS, MOLMAPS, Rcdk and Chemaxon descriptors were used to build Random Forest models for both properties. Models were subjected to the Y‐scrambling test for robustness assessment. The best models have also been validated using an external test set that is not part of the ILThermo database. A two‐phase equilibrium diagram for one of the external test set IL is presented for better visualization of the results and potential derivation of tie lines.

show abstract

Section: Introductionmentioning

confidence: 99%

QSPR Modeling of Liquid‐liquid Equilibria in Two‐phase Systems of Water and Ionic Liquid

Klimenko

Inês

Esperança

et al. 2020

Molecular Informatics

View full text Add to dashboard Cite

show abstract

“…In the work reported in our current article, we extended the work of Avdeef [ 17 ] and Klimenko et al [ 14 ] as follows. Firstly, we investigated the effect of incorporating crystallographic information, in the form of lattice energies or 3D descriptors calculated from an experimental crystal structure, into the models.…”

Section: Introductionmentioning

confidence: 97%

“…However, physics based models are not necessarily more accurate and may be more computationally expensive than QSPR approaches [ 1 , 14 ]. Interestingly, however, few QSPR models have been developed to capture the temperature dependence of solubility.…”

Section: Introductionmentioning

confidence: 99%

“…Some QSPR models were reported to predict the solubilities of organic chemicals, across a range of temperatures, in supercritical carbon dioxide for small (less than 30 chemicals), non-diverse datasets [ 15 , 16 ]. More recently, two QSPR studies sought to capture the temperature dependence of aqueous solubility for large, chemically diverse datasets [ 14 , 17 ].…”

Section: Introductionmentioning

confidence: 99%

“…In contrast, Klimenko et al [ 14 ] built a model for directly predicting aqueous solubility at a specified temperature. Their predictions were based on molecular descriptors and a descriptor derived from experimental temperature, with random forest used to train the model.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The influence of solid state information and descriptor selection on statistical models of temperature dependent aqueous solubility

Robinson

Martin

2018

J Cheminform

View full text Add to dashboard Cite

Predicting the equilibrium solubility of organic, crystalline materials at all relevant temperatures is crucial to the digital design of manufacturing unit operations in the chemical industries. The work reported in our current publication builds upon the limited number of recently published quantitative structure–property relationship studies which modelled the temperature dependence of aqueous solubility. One set of models was built to directly predict temperature dependent solubility, including for materials with no solubility data at any temperature. We propose that a modified cross-validation protocol is required to evaluate these models. Another set of models was built to predict the related enthalpy of solution term, which can be used to estimate solubility at one temperature based upon solubility data for the same material at another temperature. We investigated whether various kinds of solid state descriptors improved the models obtained with a variety of molecular descriptor combinations: lattice energies or 3D descriptors calculated from crystal structures or melting point data. We found that none of these greatly improved the best direct predictions of temperature dependent solubility or the related enthalpy of solution endpoint. This finding is surprising because the importance of the solid state contribution to both endpoints is clear. We suggest our findings may, in part, reflect limitations in the descriptors calculated from crystal structures and, more generally, the limited availability of polymorph specific data. We present curated temperature dependent solubility and enthalpy of solution datasets, integrated with molecular and crystal structures, for future investigations. Electronic supplementary materialThe online version of this article (10.1186/s13321-018-0298-3) contains supplementary material, which is available to authorized users.

show abstract

A novel strategy of coupling artificial intelligence with chemical fingerprinting to predict drug phase behaviors in complex systems

Wang,

2024

AIChE Journal

View full text Add to dashboard Cite

With the large‐scale development of drugs, understanding the drug phase behaviors in complex systems becomes more and more important. Among them, the solubility of drugs in biorelevant media needs to be urgently understood. To address this challenge, new strategies based on machine learning models are proposed. First, the strategy trains five machine learning models (extra trees [ET], gradient boosting [GB], k‐nearest neighbors [k‐NN], random forest [RF], and extreme gradient boosting [XGBoost]) based on 15 molecular descriptors of the drug molecular properties. The XGboost model was identified as the best predictive model for predicting drug solubility performance in various solvents. Next, the input feature vectors were expanded for machine learning using the MACCS chemical fingerprint coupled with the XGboost model. The MACCS chemical fingerprint coupled with XGboost model has significantly improved the prediction accuracy of drug solubility. This finding demonstrates that the proposed strategy has solubility prediction capability, which is expected to provide valid information for drug development and drug solvent screening.

show abstract

Novel enhanced applications of QSPR models: Temperature dependence of aqueous solubility

Cited by 16 publications

References 43 publications

QSPR Modeling of Liquid‐liquid Equilibria in Two‐phase Systems of Water and Ionic Liquid

QSPR Modeling of Liquid‐liquid Equilibria in Two‐phase Systems of Water and Ionic Liquid

The influence of solid state information and descriptor selection on statistical models of temperature dependent aqueous solubility

A novel strategy of coupling artificial intelligence with chemical fingerprinting to predict drug phase behaviors in complex systems

Contact Info

Product

Resources

About