Learning in silico communities to perform flow cytometric identification of synthetic bacterial communities
european conference on machine learning(2016)
摘要
Flow cytometry is able measure up to 50.000 cells in various dimensions in seconds of time. This large amount of data gives rise to the possibility of making predictions at the single-cell level, however, applied to bacterial populations a systemic investigation lacks. In order to combat this deficiency, we cultivated twenty individual bacterial populations and measured them through flow cytometry. By creating in silico communities we are able to use supervised machine learning techniques in order to examine to what extent single-cell predictions can be made; this can be used to identify the community composition. We show that for more than half of the communities consisting out of two bacterial populations we can identify single cells with an accuracy u003e90%. Furthermore we prove that in silico communities can be used to identify their in vitro counterpart communities. This result leads to the conclusion that in silico communities form a viable representation for synthetic bacterial communities, opening up new opportunities for the analysis of bacterial flow cytometric data and for the experimental study of low-complexity communities.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络