Cocrystal Prediction Using Machine Learning Models and Descriptors

Abstract

Cocrystals are of much interest in industrial application as well as academic research, and screening of suitable coformers for active pharmaceutical ingredients is the most crucial and challenging step in cocrystal development. Recently, machine learning techniques are attracting researchers in many fields including pharmaceutical research such as quantitative structure-activity/property relationship. In this paper, we develop machine learning models to predict cocrystal formation. We extract descriptor values from simplified molecular-input line-entry system (SMILES) of compounds and compare the machine learning models by experiments with our collected data of 1476 instances. As a result, we found that artificial neural network shows great potential as it has the best accuracy, sensitivity, and F1 score. We also found that the model achieved comparable performance with about half of the descriptors chosen by feature selection algorithms. We believe that this will contribute to faster and more accurate cocrystal development.

Antimalarial Drug Predictions Using Molecular Descriptors and Machine Learning against Plasmodium Falciparum

“…Quantitative structure–property relationships (QSPR) models are frequently developed using molecular descriptors, and PaDEL is amongst the attractive and well-known tools to extract descriptors [ 33 ]. There are various tools used in cheminformatics [ 31 ] such as Mordred [ 37 ], PyDPI [ 38 ], Rcpi [ 39 ], Dragon [ 40 ], and cinfony [ 41 ], which is a collection or a wrapper of other libraries such as Open Babel [ 42 ], RDKit [ 31 ] ( (accesssed on 22 June 2021)), and Chemistry Development Kit (CDK) [ 43 ]. We decided to utilize PaDEL because of its advantages: it provides approximately 1875 molecular descriptors within a brief execution time, and it is simple to install and utilize.…”

“…We have implemented various ML models such as artificial neural network (ANN), support vector machine (SVM) [ 46 ], random forest (RF) [ 47 ], extreme gradient boost (XGB) [ 48 ], and Logistic Regression (LR) [ 49 ]. The ANN is recognized to be useful in a variety of research fields, including image analysis, natural language processing, and speech recognition; if it has a deep structure, it is a deep learning model (i.e., multiple hidden layers) [ 31 ]. The SVM is known to be successful in many classification applications and tasks [ 50 ], and it identifies a decision boundary based on boundary examples or instances (i.e., support vectors).…”

“…The SVM is known to be successful in many classification applications and tasks [ 50 ], and it identifies a decision boundary based on boundary examples or instances (i.e., support vectors). The RF and XGB are both standard and common ensemble techniques, although the RF employs a bagging strategy while the XGB uses a boosting strategy [ 31 ]. The LR, a model with the sigmoid function often utilized by statisticians to describe properties of population growth in ecology, is rising quickly and maxing out at the carrying capacity of the environment.…”

“…Two feature selection algorithms are employed: Recursive Feature Elimination (RFE) and K-best algorithm. The K-best is a filter-based algorithm that selects potential features according to a particular function , where f and c are a feature and a label, respectively, while the RFE is a wrapper-based algorithm that treats the feature selection as a search problem [ 31 ], and eliminates unpromising features on a regular basis until only the desired number of features remains. The ANN model was used as an estimator of the RFE algorithm and took the ANOVA F-value as the function .…”

“…For example, Ref. [ 31 ] addressed the major crucial and critical fundamental problems (i.e., poor solubility, bioavailability, and efficacy of drugs) that hinder the drug development process through improving specific physicochemical and biopharmaceutical properties of active pharmaceutical ingredients (APIs), by applying ML models to predict which pair of API and coformer will successfully result in the new cocrystal formation that eventually becomes new drug and medicine after the Food and Drug Administration (FDA) approval, from a set of chemical experiments between API and the coformer since the essential and difficult phase in cocrystal production as an auxiliary state-of-the-art form to boost and enhance drug development is the screening of suitable coformers for an API. Danishuddin et al [ 9 ] established the development and rigorous validation of antimalarial predictive models using machine learning approaches and ultimately achieved an accuracy of ∼85.00%.…”