Deductive automated pollen classification in environmental samples via exploratory deep learning and imaging flow cytometry.

Pollen and tracheophyte spores are ubiquitous environmental indicators at local and global scales. Palynology is typically performed manually by microscopic analysis; a specialised and time-consuming task limited in taxonomical precision and sampling frequency, therefore restricting data quality used to inform climate change and pollen forecasting models. We build on the growing work using AI (artificial intelligence) for automated pollen classification to design a flexible network that can deal with the uncertainty of broad-scale environmental applications. We combined imaging flow cytometry with Guided Deep Learning to identify and accurately categorise pollen in environmental samples; here, pollen grains captured within c. 5500 Cal yr BP old lake sediments. Our network discriminates not only pollen included in training libraries to the species level but, depending on the sample, can classify previously unseen pollen to the likely phylogenetic order, family and even genus. Our approach offers valuable insights into the development of a widely transferable, rapid and accurate exploratory tool for pollen classification in 'real-world' environmental samples with improved accuracy over pure deep learning techniques. This work has the potential to revolutionise many aspects of palynology, allowing a more detailed spatial and temporal understanding of pollen in the environment with improved taxonomical resolution.

A bacterial process for selenium nanosphere assembly.

During selenate respiration by Thauera selenatis, the reduction of selenate results in the formation of intracellular selenium (Se) deposits that are ultimately secreted as Se nanospheres of approximately 150 nm in diameter. We report that the Se nanospheres are associated with a protein of approximately 95 kDa. Subsequent experiments to investigate the expression and secretion profile of this protein have demonstrated that it is up-regulated and secreted in response to increasing selenite concentrations. The protein was purified from Se nanospheres, and peptide fragments from a tryptic digest were used to identify the gene in the draft T. selenatis genome. A matched open reading frame was located, encoding a protein with a calculated mass of 94.5 kDa. N-terminal sequence analysis of the mature protein revealed no cleavable signal peptide, suggesting that the protein is exported directly from the cytoplasm. The protein has been called Se factor A (SefA), and homologues of known function have not been reported previously. The sefA gene was cloned and expressed in Escherichia coli, and the recombinant His-tagged SefA purified. In vivo experiments demonstrate that SefA forms larger (approximately 300 nm) Se nanospheres in E. coli when treated with selenite, and these are retained within the cell. In vitro assays demonstrate that the formation of Se nanospheres upon the reduction of selenite by glutathione are stabilized by the presence of SefA. The role of SefA in selenium nanosphere assembly has potential for exploitation in bionanomaterial fabrication.