Predicting Drug Solubility Using Different Machine Learning Methods -- Linear Regression Model with Extracted Chemical Features vs Graph Convolutional Neural Network

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ho, John, Yin, Zhao-Heng, Zhang, Colin, Guo, Nicole, Ha, Yang
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910287642230784
author Ho, John
Yin, Zhao-Heng
Zhang, Colin
Guo, Nicole
Ha, Yang
author_facet Ho, John
Yin, Zhao-Heng
Zhang, Colin
Guo, Nicole
Ha, Yang
contents Predicting the solubility of given molecules remains crucial in the pharmaceutical industry. In this study, we revisited this extensively studied topic, leveraging the capabilities of contemporary computing resources. We employed two machine learning models: a linear regression model and a graph convolutional neural network (GCNN) model, using various experimental datasets. Both methods yielded reasonable predictions, with the GCNN model exhibiting the highest level of performance. However, the present GCNN model has limited interpretability while the linear regression model allows scientists for a greater in-depth analysis of the underlying factors through feature importance analysis, although more human inputs and evaluations on the overall dataset is required. From the perspective of chemistry, using the linear regression model, we elucidated the impact of individual atom species and functional groups on overall solubility, highlighting the significance of comprehending how chemical structure influences chemical properties in the drug development process. It is learned that introducing oxygen atoms can increase the solubility of organic molecules, while almost all other hetero atoms except oxygen and nitrogen tend to decrease solubility.
format Preprint
id arxiv_https___arxiv_org_abs_2308_12325
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Predicting Drug Solubility Using Different Machine Learning Methods -- Linear Regression Model with Extracted Chemical Features vs Graph Convolutional Neural Network
Ho, John
Yin, Zhao-Heng
Zhang, Colin
Guo, Nicole
Ha, Yang
Quantitative Methods
Machine Learning
Predicting the solubility of given molecules remains crucial in the pharmaceutical industry. In this study, we revisited this extensively studied topic, leveraging the capabilities of contemporary computing resources. We employed two machine learning models: a linear regression model and a graph convolutional neural network (GCNN) model, using various experimental datasets. Both methods yielded reasonable predictions, with the GCNN model exhibiting the highest level of performance. However, the present GCNN model has limited interpretability while the linear regression model allows scientists for a greater in-depth analysis of the underlying factors through feature importance analysis, although more human inputs and evaluations on the overall dataset is required. From the perspective of chemistry, using the linear regression model, we elucidated the impact of individual atom species and functional groups on overall solubility, highlighting the significance of comprehending how chemical structure influences chemical properties in the drug development process. It is learned that introducing oxygen atoms can increase the solubility of organic molecules, while almost all other hetero atoms except oxygen and nitrogen tend to decrease solubility.
title Predicting Drug Solubility Using Different Machine Learning Methods -- Linear Regression Model with Extracted Chemical Features vs Graph Convolutional Neural Network
topic Quantitative Methods
Machine Learning
url https://arxiv.org/abs/2308.12325