FormulationBCS: A Machine Learning Platform Based on Diverse Molecular Representations for Biopharmaceutical Classification System (BCS) Class Prediction.

IF 4.5 2区 医学 Q2 MEDICINE, RESEARCH & EXPERIMENTAL
Molecular Pharmaceutics Pub Date : 2025-01-06 Epub Date: 2024-12-08 DOI:10.1021/acs.molpharmaceut.4c00946
Zheng Wu, Nannan Wang, Zhuyifan Ye, Huanle Xu, Ging Chan, Defang Ouyang
{"title":"FormulationBCS: A Machine Learning Platform Based on Diverse Molecular Representations for Biopharmaceutical Classification System (BCS) Class Prediction.","authors":"Zheng Wu, Nannan Wang, Zhuyifan Ye, Huanle Xu, Ging Chan, Defang Ouyang","doi":"10.1021/acs.molpharmaceut.4c00946","DOIUrl":null,"url":null,"abstract":"<p><p>The Biopharmaceutics Classification System (BCS) has facilitated biowaivers and played a significant role in enhancing drug regulation and development efficiency. However, the productivity of measuring the key discriminative properties of BCS, solubility and permeability, still requires improvement, limiting high-throughput applications of BCS, which is essential for evaluating drug candidate developability and guiding formulation decisions in the early stages of drug development. In recent years, advancements in machine learning (ML) and molecular characterization have revealed the potential of quantitative structure-performance relationships (QSPR) for rapid and accurate <i>in silico</i> BCS classification. The present study aims to develop a web platform for high-throughput BCS classification based on high-performance ML models. Initially, four data sets of BCS-related molecular properties: log <i>S</i>, log <i>P</i>, log <i>D</i>, and log <i>P</i><sub>app</sub> were curated. Subsequently, 6 ML algorithms or deep learning frameworks were employed to construct models, with diverse molecular representations ranging from one-dimensional molecular fingerprints, descriptors, and molecular graphs to three-dimensional molecular spatial coordinates. By comparing different combinations of molecular representations and learning algorithms, LightGBM exhibited excellent performance in solubility prediction, with an <i>R</i><sup>2</sup> of 0.84; AttentiveFP outperformed others in permeability prediction, with <i>R</i><sup>2</sup> values of 0.96 and 0.76 for log <i>P</i> and log <i>D</i>, respectively; and XGBoost was the most accurate for log <i>P</i><sub>app</sub> prediction, with an <i>R</i><sup>2</sup> of 0.71. When externally validated on a marketed drug BCS category data set, the best-performing models achieved classification accuracies of over 77 and 73% for solubility and permeability, respectively. Finally, the well-trained models were embedded into the first ML-based BCS class prediction web platform (x f), enabling pharmaceutical scientists to quickly determine the BCS category of candidate drugs, which will aid in the high-throughput BCS assessment for candidate drugs during the preformulation stage, thereby promoting reduced risk and enhanced efficiency in drug development and regulation.</p>","PeriodicalId":52,"journal":{"name":"Molecular Pharmaceutics","volume":" ","pages":"330-342"},"PeriodicalIF":4.5000,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11707745/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Molecular Pharmaceutics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1021/acs.molpharmaceut.4c00946","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/12/8 0:00:00","PubModel":"Epub","JCR":"Q2","JCRName":"MEDICINE, RESEARCH & EXPERIMENTAL","Score":null,"Total":0}
引用次数: 0

Abstract

The Biopharmaceutics Classification System (BCS) has facilitated biowaivers and played a significant role in enhancing drug regulation and development efficiency. However, the productivity of measuring the key discriminative properties of BCS, solubility and permeability, still requires improvement, limiting high-throughput applications of BCS, which is essential for evaluating drug candidate developability and guiding formulation decisions in the early stages of drug development. In recent years, advancements in machine learning (ML) and molecular characterization have revealed the potential of quantitative structure-performance relationships (QSPR) for rapid and accurate in silico BCS classification. The present study aims to develop a web platform for high-throughput BCS classification based on high-performance ML models. Initially, four data sets of BCS-related molecular properties: log S, log P, log D, and log Papp were curated. Subsequently, 6 ML algorithms or deep learning frameworks were employed to construct models, with diverse molecular representations ranging from one-dimensional molecular fingerprints, descriptors, and molecular graphs to three-dimensional molecular spatial coordinates. By comparing different combinations of molecular representations and learning algorithms, LightGBM exhibited excellent performance in solubility prediction, with an R2 of 0.84; AttentiveFP outperformed others in permeability prediction, with R2 values of 0.96 and 0.76 for log P and log D, respectively; and XGBoost was the most accurate for log Papp prediction, with an R2 of 0.71. When externally validated on a marketed drug BCS category data set, the best-performing models achieved classification accuracies of over 77 and 73% for solubility and permeability, respectively. Finally, the well-trained models were embedded into the first ML-based BCS class prediction web platform (x f), enabling pharmaceutical scientists to quickly determine the BCS category of candidate drugs, which will aid in the high-throughput BCS assessment for candidate drugs during the preformulation stage, thereby promoting reduced risk and enhanced efficiency in drug development and regulation.

BCS:生物制药分类系统(BCS)分类预测中基于不同分子表示的机器学习平台。
生物制药分类系统(BCS)为生物豁免提供了便利,在提高药物监管和开发效率方面发挥了重要作用。然而,测量BCS关键鉴别性质(溶解度和渗透性)的效率仍有待提高,这限制了BCS的高通量应用,而这对于评估候选药物的可开发性和指导药物开发早期的处方决策至关重要。近年来,机器学习(ML)和分子表征的进步揭示了定量结构-性能关系(QSPR)在快速准确的硅BCS分类中的潜力。本研究旨在开发一个基于高性能机器学习模型的高通量BCS分类web平台。最初,bcs相关分子特性的四组数据集:log S、log P、log D和log Papp。随后,采用6 ML算法或深度学习框架构建模型,具有从一维分子指纹、描述符、分子图到三维分子空间坐标等多种分子表征。通过比较不同的分子表示和学习算法组合,LightGBM在溶解度预测方面表现优异,R2为0.84;AttentiveFP在渗透率预测方面优于其他方法,对数P和对数D的R2分别为0.96和0.76;XGBoost对log Papp预测最准确,R2为0.71。当在上市药物BCS类别数据集上进行外部验证时,表现最好的模型在溶解度和渗透率方面的分类准确率分别超过77%和73%。最后,将训练有素的模型嵌入到第一个基于ml的BCS类别预测网络平台(x f)中,使制药科学家能够快速确定候选药物的BCS类别,这将有助于在预制剂阶段对候选药物进行高通量BCS评估,从而降低药物开发和监管的风险,提高效率。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Molecular Pharmaceutics
Molecular Pharmaceutics 医学-药学
CiteScore
8.00
自引率
6.10%
发文量
391
审稿时长
2 months
期刊介绍: Molecular Pharmaceutics publishes the results of original research that contributes significantly to the molecular mechanistic understanding of drug delivery and drug delivery systems. The journal encourages contributions describing research at the interface of drug discovery and drug development. Scientific areas within the scope of the journal include physical and pharmaceutical chemistry, biochemistry and biophysics, molecular and cellular biology, and polymer and materials science as they relate to drug and drug delivery system efficacy. Mechanistic Drug Delivery and Drug Targeting research on modulating activity and efficacy of a drug or drug product is within the scope of Molecular Pharmaceutics. Theoretical and experimental peer-reviewed research articles, communications, reviews, and perspectives are welcomed.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信