{"title":"PEZy-miner: An artificial intelligence driven approach for the discovery of plastic-degrading enzyme candidates","authors":"","doi":"10.1016/j.mec.2024.e00248","DOIUrl":null,"url":null,"abstract":"<div><p>Plastic waste has caused a global environmental crisis. Biocatalytic depolymerization mediated by enzymes has emerged as an efficient and sustainable alternative for plastic treatment and recycling. However, it is challenging and time-consuming to discover novel plastic-degrading enzymes using conventional cultivation-based or omics methods. There is a growing interest in developing effective computational methods to identify new enzymes with desirable plastic degradation functionalities by exploring the ever-increasing databases of protein sequences. In this study, we designed an innovative machine learning-based framework, named PEZy-Miner, to mine for enzymes with high potential in degrading plastics of interest. Two datasets integrating information from experimentally verified enzymes and homologs with unknown plastic-degrading activity were created respectively, covering eleven types of plastic substrates. Protein language models and binary classification models were developed to predict enzymatic degradation of plastics along with confidence and uncertainty estimation. PEZy-Miner exhibited high prediction accuracy and stability when validated on experimentally verified enzymes. Furthermore, by masking the experimentally verified enzymes and blending them into homolog dataset, PEZy-Miner effectively concentrated the experimentally verified entries by 14∼30 times while shortlisting promising plastic-degrading enzyme candidates. We applied PEZy-Miner to 0.1 million putative sequences, out of which 27 new sequences were identified with high confidence. This study provided a new computational tool for mining and recommending promising new plastic-degrading enzymes.</p></div>","PeriodicalId":18695,"journal":{"name":"Metabolic Engineering Communications","volume":null,"pages":null},"PeriodicalIF":3.7000,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2214030124000178/pdfft?md5=a6ab15db96315a11ed6d106b7d4eb890&pid=1-s2.0-S2214030124000178-main.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Metabolic Engineering Communications","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2214030124000178","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"BIOTECHNOLOGY & APPLIED MICROBIOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Plastic waste has caused a global environmental crisis. Biocatalytic depolymerization mediated by enzymes has emerged as an efficient and sustainable alternative for plastic treatment and recycling. However, it is challenging and time-consuming to discover novel plastic-degrading enzymes using conventional cultivation-based or omics methods. There is a growing interest in developing effective computational methods to identify new enzymes with desirable plastic degradation functionalities by exploring the ever-increasing databases of protein sequences. In this study, we designed an innovative machine learning-based framework, named PEZy-Miner, to mine for enzymes with high potential in degrading plastics of interest. Two datasets integrating information from experimentally verified enzymes and homologs with unknown plastic-degrading activity were created respectively, covering eleven types of plastic substrates. Protein language models and binary classification models were developed to predict enzymatic degradation of plastics along with confidence and uncertainty estimation. PEZy-Miner exhibited high prediction accuracy and stability when validated on experimentally verified enzymes. Furthermore, by masking the experimentally verified enzymes and blending them into homolog dataset, PEZy-Miner effectively concentrated the experimentally verified entries by 14∼30 times while shortlisting promising plastic-degrading enzyme candidates. We applied PEZy-Miner to 0.1 million putative sequences, out of which 27 new sequences were identified with high confidence. This study provided a new computational tool for mining and recommending promising new plastic-degrading enzymes.
期刊介绍:
Metabolic Engineering Communications, a companion title to Metabolic Engineering (MBE), is devoted to publishing original research in the areas of metabolic engineering, synthetic biology, computational biology and systems biology for problems related to metabolism and the engineering of metabolism for the production of fuels, chemicals, and pharmaceuticals. The journal will carry articles on the design, construction, and analysis of biological systems ranging from pathway components to biological complexes and genomes (including genomic, analytical and bioinformatics methods) in suitable host cells to allow them to produce novel compounds of industrial and medical interest. Demonstrations of regulatory designs and synthetic circuits that alter the performance of biochemical pathways and cellular processes will also be presented. Metabolic Engineering Communications complements MBE by publishing articles that are either shorter than those published in the full journal, or which describe key elements of larger metabolic engineering efforts.