開發運用人工智慧篩選可降解新興環境有機污染物如阻燃劑微生物之新穎技術
Develop a novel methodology to screen microbial degraders for emerging contaminants such as flame retardants by artificial intelligence
國立臺灣大學農化系 施養信 教授
中央研究院資訊所 林仲彥 研究員
計畫執行期間：2021.1.1 – 2022.12.31
優良研究成果
Recent advancements in sequencing technology have revolutionized our understanding of the microbiome, allowing for deeper insights into microbial communities and their functional genes. This study utilized machine learning techniques to identify degraders of environmental contaminants, specifically flame retardants, from microbial genomes and proteomes databases. Through alignment of protein sequences, a promising degrader, Ralstonia solanacearum, was identified. This bacterium exhibited significant effects on the removal of the flame retardant HBCD, degrading 41% within 12 days and 69% within 36 days of incubation.
Additionally, soil microbial community composition was observed to alter in response to HBCD treatment, with novel bacterial taxa identified to contribute to its biotransformation (publication 1). Leveraging these findings, the study developed a predictive model using machine learning algorithms, achieving promising accuracy and reliability in predicting HBCD degraders. This integrated approach, combining metagenomic analysis and machine learning, not only advances our understanding of HBCD biotransformation in natural environments but also provides a efficient screening method. The model developed in this study holds promise for applications beyond HBCD, potentially extending to other emerging flame retardants and environmental pollutants.
Publication:
Yi-Jie Li, Chia-Hsien Chuang, Wen-Chih Cheng, Shu-Hwa Chen, Wen-Ling Chen, Yu-Jie Lin, Chung-Yen Lin*, Yang-hsin Shih*, “A metagenomics study of hexabromocyclododecane degradation with a soil microbial community,” Journal of Hazardous Materials, volume 430, pages 128465, May 2022, (IF=14.24)