期刊
JOURNAL OF MICROBIOLOGY
卷 59, 期 3, 页码 259-269出版社
MICROBIOLOGICAL SOCIETY KOREA
DOI: 10.1007/s12275-021-0668-9
关键词
fecal pollution; microbial source tracking; metagenomics; machine learning; next generation sequencing
类别
资金
- Korea Disease Control and Prevention Agency [2020ER5408-00]
- Minnesota Agricultural Experiment Station
- Basic Science Research Program to Research Institute for Basic Sciences (RIBS) of Jeju National University through the National Research Foundation of Korea (NRF) - Ministry of Education [2019R1A6A1A10072987]
Despite the challenges in identifying sources of fecal bacteria in complex environments, advancements in qPCR and next generation sequencing technologies have shifted microbial source tracking towards culture independent methods. Metagenomic tools can provide deep insight into host specific fecal markers, while machine learning algorithms can offer statistically robust and automated platforms for optimization of resources. The successful application of machine learning models in various fields suggests that they may serve as a better alternative to traditional microbial source tracking approaches in the future.
The environment is under siege from a variety of pollution sources. Fecal pollution is especially harmful as it disperses pathogenic bacteria into waterways. Unraveling origins of mixed sources of fecal bacteria is difficult and microbial source tracking (MST) in complex environments is still a daunting task. Despite the challenges, the need for answers far outweighs the difficulties experienced. Advancements in qPCR and next generation sequencing (NGS) technologies have shifted the traditional culture-based MST approaches towards culture independent technologies, where community-based MST is becoming a method of choice. Metagenomic tools may be useful to overcome some of the limitations of community-based MST methods as they can give deep insight into identifying host specific fecal markers and their association with different environments. Adoption of machine learning (ML) algorithms, along with the metagenomic based MST approaches, will also provide a statistically robust and automated platform. To compliment that, ML-based approaches provide accurate optimization of resources. With the successful application of ML based models in disease prediction, outbreak investigation and medicine prescription, it would be possible that these methods would serve as a better surrogate of traditional MST approaches in future.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据