Pushing the limits of solubility prediction via quality-oriented data selection

Sorkun, Murat Cihan; Koelman, J.M.V.A.; Er, Süleyman

doi:10.1016/j.isci.2020.101961

Cited by 44 publications

(36 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The AqSolPred model, which was used for solubility predictions in the current work, had previously been validated on a benchmark solubility dataset 24 . The model has a Mean Absolute Error of 0.348 LogS, which is lower than the conventional cheminformatics and ML methods that are ordinarily used for the prediction of aqueous solubility of chemical species 13 .…”

Section: Validation Of Solubility Predictionsmentioning

confidence: 89%

“…It is an exemplary resource on quinone and aza-aromatic electroactive compounds as it contains several candidate molecules for batteries that are worthy of experimental investigation. The database contains comprehensive data that has been systematically collected by using the state-of-the-art computational procedures 1112 and data-driven methods 13 . Therefore, it's also useful for other applications beyond ARFBs for which the intriguing chemistry of these molecules matter.…”

Section: Background and Summarymentioning

confidence: 99%

“…The water solubility data of the compounds was built by using the Aqueous Solubility Prediction Model (AqSolPred v1.0) 13 . AqSolPred is a supervised and consensus ML model that was empowered by training on a large, curated, and reference aqueous solubility database, AqSolDB 22 .…”

Section: B Solubility Predictionsmentioning

confidence: 99%

See 2 more Smart Citations

RedDB, a Computational Database of Electroactive Molecules for Aqueous Redox Flow Batteries

Sorkun

Zhang²,

Khetan³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

An increasing number of electroactive compounds have recently been explored for their use in high-performance redox flow batteries for grid-scale energy storage. Given the vast and highly diverse chemical space of the candidate compounds, it is alluring to access their physicochemical properties in a speedy way. High-throughput virtual screening approaches, which use powerful combinatorial techniques for systematic enumerations of large virtual chemical libraries and respective property evaluations, are indispensable tools for an agile exploration of the designated chemical space. Herein, RedDB: a computational database that contains 31,677 molecules from two prominent classes of organic electroactive compounds, quinones and aza-aromatics, has been presented. RedDB incorporates miscellaneous physicochemical property information of the compounds that can potentially be employed as battery performance descriptors. RedDB’s development steps, including: i)chemical library generation, ii) molecular property prediction based on quantum chemical calculations, iii) aqueous solubility prediction using machine learning, and iv) data processing and database creation, have been described.

show abstract

Section: Validation Of Solubility Predictionsmentioning

confidence: 89%

Section: Background and Summarymentioning

confidence: 99%

See 1 more Smart Citation

RedDB, a Computational Database of Electroactive Molecules for Aqueous Redox Flow Batteries

Sorkun

Zhang²,

Khetan³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…If it is true that the quality of QSPR models is directly influenced by the quality of the datasets on which they are trained, it is equally true that the assessment of predictive performances of a model is strongly influenced by the accuracy of the data on which it is tested. Indeed, the observed performances derive from the contribution of the actual predictive performances (defined as the accuracy of a model that would be observed on a test set with zero internal error) and the uncertainty in the test data [ 38 ]. In the case of the intrinsic solubility, due to poor reliability of experimental solubility values, the observed performances could be significantly influenced by the errors affecting the logS 0 values of the test compounds.…”

Section: Discussionmentioning

confidence: 99%

Application of Artificial Neural Networks to Predict the Intrinsic Solubility of Drug-Like Molecules

2021

View full text Add to dashboard Cite

Machine learning (ML) approaches are receiving increasing attention from pharmaceutical companies and regulatory agencies, given their ability to mine knowledge from available data. In drug discovery, for example, they are employed in quantitative structure–property relationship (QSPR) models to predict biological properties from the chemical structure of a drug molecule. In this paper, following the Second Solubility Challenge (SC-2), a QSPR model based on artificial neural networks (ANNs) was built to predict the intrinsic solubility (logS0) of the 100-compound low-variance tight set and the 32-compound high-variance loose set provided by SC-2 as test datasets. First, a training dataset of 270 drug-like molecules with logS0 value experimentally determined was gathered from the literature. Then, a standard three-layer feed-forward neural network was defined by using 10 ChemGPS physico-chemical descriptors as input features. The developed ANN showed adequate predictive performances on both of the SC-2 test datasets. Benefits and limitations of ML approaches have been highlighted and discussed, starting from this case-study. The main findings confirmed that ML approaches are an attractive and promising tool to predict logS0; however, many aspects, such as data quality, molecular descriptor computation and selection, and assessment of applicability domain, are crucial but often neglected, and should be carefully considered to improve predictions based on ML.

show abstract

“…29,30 Accordingly, the linear regression equations used for predicting the experimental redox potentials of the reduction of C=O groups on quinone-like 29 The aqueous solubility data of the reactant molecules was obtained by using AqSolPred v1.0. 31 , which is a state-of-the-art consensus ML model that had been trained on the largest open-source measured solubility dataset of chemical compounds. 20 In this context, the approach employed here is in stark contrast to other HTVS efforts that make use of the computed aqueous solvation free energy, ∆ 8"9: " , as an approximate descriptor for the solubility of compounds.…”

Section: Physics-based and Data-driven Modellingmentioning

confidence: 99%

Data-driven discovery of small electroactive molecules for energy storage in aqueous redox flow batteries

Zhang

Khetan

Sorkun

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Owing to advancements in computing power and improvements in automation of modelling, high-throughput virtual screening (HTVS) has increasingly been used for materials data generation.Here, we applied a HTVS-guided experimental study for the large-scale exploration of quinone-like anolytes for aqueous redox flow batteries (ARFBs). This includes the design of a focused virtual chemical library inspired by small colorant molecules, quantum chemical prediction of redox properties, machine learning prediction of aqueous solubility, automated search for commercial availability on vendor databases, and electrochemical characterization of the most promising compounds. Screening efforts in a chemical space of 3,257 redox pairs led to 205 predicted candidates with higher solubility and lower redox potential than that of the state-of-the-art anthraquinone-2,7disulfonic acid (AQDS) anolyte used in ARFBs. Through the electrochemical studies on the commercially available compounds, we identified the molecules that show good performance in an ARFB setup. Among them, indigo trisulfonate [Indigo-3(SO3H)] showed higher solubility, capacity retention, and coulombic efficiency than AQDS and its predecessors. The data-driven material design methodology presented here is flexible and applicable for the future exploration of small compounds for electrochemical energy storage.

show abstract

Pushing the limits of solubility prediction via quality-oriented data selection

Cited by 44 publications

References 38 publications

RedDB, a Computational Database of Electroactive Molecules for Aqueous Redox Flow Batteries

RedDB, a Computational Database of Electroactive Molecules for Aqueous Redox Flow Batteries

Application of Artificial Neural Networks to Predict the Intrinsic Solubility of Drug-Like Molecules

Data-driven discovery of small electroactive molecules for energy storage in aqueous redox flow batteries

Contact Info

Product

Resources

About