Vinamr Jain, Michael T. Bergman, Carol K. Hall, Fengqi You
{"title":"人工智能驱动的混杂和选择性塑料结合肽的合理设计","authors":"Vinamr Jain, Michael T. Bergman, Carol K. Hall, Fengqi You","doi":"10.1039/d5sc04903b","DOIUrl":null,"url":null,"abstract":"Microplastic pollution is challenging to remediate due to the small size and heterogeneous composition of microplastic particles. Remediation efforts would benefit from tools that either bind to the many components of microplastic pollution (promiscuous binding) to facilitate quantitation and capture, or bind to certain components of pollution (selective binding) to facilitate separation or degradation. Such a role could be filled by polypeptides, which can bind selectively or promiscuously to biomolecules or materials. While methods exist to design plastic-binding peptides (PBPs) for a single plastic, the design of promiscuous plastic-binding peptides has received scant attention, and there are no methods to design selective plastic-binding peptides. Here, we present a minimalist yet high-performing framework integrating Long Short-Term Memory (LSTM) models with simulated annealing (SA) to design promiscuous plastic-binding or selective plastic-binding peptides. Our approach learns sequence–function relationships governing peptide affinity for different plastics from PepBD data, a biophysical modeling program. The learned relationship enables rapid design of peptides with tailored binding properties for arbitrary combinations of plastics. We use our LSTM-SA framework to engineer (1) promiscuous plastic-binding peptides with affinity for five plastics (polyethylene, polypropylene, PET, polyvinyl chloride, and nylon), and (2) selective plastic-binding peptides that bind preferentially to one plastic (polypropylene) over another (PET). Notably, the promiscuous plastic-binding peptides are the first reported designs intended to bind to nylon and PVC. Molecular dynamics simulations validate that our designed peptides exhibit the predicted binding behaviors, where high affinity is linked to strong van der Waals interactions. The framework's modularity can be readily adapted to optimize peptide selectivity or promiscuity for different combinations of plastics. More broadly, the architecture may be useful for designing peptides that bind to other solid materials.","PeriodicalId":9909,"journal":{"name":"Chemical Science","volume":"1 1","pages":""},"PeriodicalIF":7.4000,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"AI-driven rational design of promiscuous and selective plastic-binding peptides\",\"authors\":\"Vinamr Jain, Michael T. Bergman, Carol K. Hall, Fengqi You\",\"doi\":\"10.1039/d5sc04903b\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Microplastic pollution is challenging to remediate due to the small size and heterogeneous composition of microplastic particles. Remediation efforts would benefit from tools that either bind to the many components of microplastic pollution (promiscuous binding) to facilitate quantitation and capture, or bind to certain components of pollution (selective binding) to facilitate separation or degradation. Such a role could be filled by polypeptides, which can bind selectively or promiscuously to biomolecules or materials. While methods exist to design plastic-binding peptides (PBPs) for a single plastic, the design of promiscuous plastic-binding peptides has received scant attention, and there are no methods to design selective plastic-binding peptides. Here, we present a minimalist yet high-performing framework integrating Long Short-Term Memory (LSTM) models with simulated annealing (SA) to design promiscuous plastic-binding or selective plastic-binding peptides. Our approach learns sequence–function relationships governing peptide affinity for different plastics from PepBD data, a biophysical modeling program. The learned relationship enables rapid design of peptides with tailored binding properties for arbitrary combinations of plastics. We use our LSTM-SA framework to engineer (1) promiscuous plastic-binding peptides with affinity for five plastics (polyethylene, polypropylene, PET, polyvinyl chloride, and nylon), and (2) selective plastic-binding peptides that bind preferentially to one plastic (polypropylene) over another (PET). Notably, the promiscuous plastic-binding peptides are the first reported designs intended to bind to nylon and PVC. Molecular dynamics simulations validate that our designed peptides exhibit the predicted binding behaviors, where high affinity is linked to strong van der Waals interactions. The framework's modularity can be readily adapted to optimize peptide selectivity or promiscuity for different combinations of plastics. More broadly, the architecture may be useful for designing peptides that bind to other solid materials.\",\"PeriodicalId\":9909,\"journal\":{\"name\":\"Chemical Science\",\"volume\":\"1 1\",\"pages\":\"\"},\"PeriodicalIF\":7.4000,\"publicationDate\":\"2025-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Chemical Science\",\"FirstCategoryId\":\"92\",\"ListUrlMain\":\"https://doi.org/10.1039/d5sc04903b\",\"RegionNum\":1,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CHEMISTRY, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Chemical Science","FirstCategoryId":"92","ListUrlMain":"https://doi.org/10.1039/d5sc04903b","RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}
AI-driven rational design of promiscuous and selective plastic-binding peptides
Microplastic pollution is challenging to remediate due to the small size and heterogeneous composition of microplastic particles. Remediation efforts would benefit from tools that either bind to the many components of microplastic pollution (promiscuous binding) to facilitate quantitation and capture, or bind to certain components of pollution (selective binding) to facilitate separation or degradation. Such a role could be filled by polypeptides, which can bind selectively or promiscuously to biomolecules or materials. While methods exist to design plastic-binding peptides (PBPs) for a single plastic, the design of promiscuous plastic-binding peptides has received scant attention, and there are no methods to design selective plastic-binding peptides. Here, we present a minimalist yet high-performing framework integrating Long Short-Term Memory (LSTM) models with simulated annealing (SA) to design promiscuous plastic-binding or selective plastic-binding peptides. Our approach learns sequence–function relationships governing peptide affinity for different plastics from PepBD data, a biophysical modeling program. The learned relationship enables rapid design of peptides with tailored binding properties for arbitrary combinations of plastics. We use our LSTM-SA framework to engineer (1) promiscuous plastic-binding peptides with affinity for five plastics (polyethylene, polypropylene, PET, polyvinyl chloride, and nylon), and (2) selective plastic-binding peptides that bind preferentially to one plastic (polypropylene) over another (PET). Notably, the promiscuous plastic-binding peptides are the first reported designs intended to bind to nylon and PVC. Molecular dynamics simulations validate that our designed peptides exhibit the predicted binding behaviors, where high affinity is linked to strong van der Waals interactions. The framework's modularity can be readily adapted to optimize peptide selectivity or promiscuity for different combinations of plastics. More broadly, the architecture may be useful for designing peptides that bind to other solid materials.
期刊介绍:
Chemical Science is a journal that encompasses various disciplines within the chemical sciences. Its scope includes publishing ground-breaking research with significant implications for its respective field, as well as appealing to a wider audience in related areas. To be considered for publication, articles must showcase innovative and original advances in their field of study and be presented in a manner that is understandable to scientists from diverse backgrounds. However, the journal generally does not publish highly specialized research.