RANdom SAmple Consensus (RANSAC) algorithm for material-informatics: application to photovoltaic solar cells

• Published in: Journal of cheminformatics Vol. 9; no. 1; p. 34
• Main Authors: Kaspi, Omer, Yosipof, Abraham, Senderowitz, Hanoch
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 06.06.2017; BioMed Central Ltd; Springer Nature B.V; BMC
• Subjects: Algorithms; Cell culture; Chemical research; Chemistry; Chemistry and Materials Science; Computational Biology/Bioinformatics; Computational chemistry; Computer Applications in Chemistry; Datasets; Documentation and Information in Chemistry; Image processing; Informatics; Learning algorithms; Libraries; Machine learning; Material-informatics; Mathematical models; Photovoltaic cells; Photovoltaic power generation; Photovoltaics; Prediction models; Properties; Properties (attributes); QSAR; RANSAC; Research Article; Samples; Solar Cells; Statistical analysis; Statistical methods; Structure-activity relationships; Theoretical and Computational Chemistry; RANSAC; Material-informatics; QSAR; Solar Cells; Photovoltaics
• ISSN: 1758-2946; 1758-2946
• DOI: 10.1186/s13321-017-0224-0

An important aspect of chemoinformatics and material-informatics is the usage of machine learning algorithms to build Quantitative Structure Activity Relationship (QSAR) models. The RANdom SAmple Consensus (RANSAC) algorithm is a predictive modeling tool widely used in the image processing field for cleaning datasets from noise. RANSAC could be used as a “one stop shop” algorithm for developing and validating QSAR models, performing outlier removal, descriptors selection, model development and predictions for test set samples using applicability domain. For “future” predictions (i.e., for samples not included in the original test set) RANSAC provides a statistical estimate for the probability of obtaining reliable predictions, i.e., predictions within a pre-defined number of standard deviations from the true values. In this work we describe the first application of RNASAC in material informatics, focusing on the analysis of solar cells. We demonstrate that for three datasets representing different metal oxide (MO) based solar cell libraries RANSAC-derived models select descriptors previously shown to correlate with key photovoltaic properties and lead to good predictive statistics for these properties. These models were subsequently used to predict the properties of virtual solar cells libraries highlighting interesting dependencies of PV properties on MO compositions.