Deep metric learning improves lab of origin prediction of genetically engineered plasmids

Genome engineering is undergoing unprecedented development and is now becoming widely available. To ensure responsible biotechnology innovation and to reduce misuse of engineered DNA sequences, it is vital to develop tools to identify the lab-of-origin of engineered plasmids. Genetic engineering attribution (GEA), the ability to make sequence-lab associations, would supportforensic experts in this process. Here, we propose a method, based on metric learning, that ranks the most likely labs-of-origin whilstsimultaneously generating embeddings for plasmid sequences and labs. These embeddings can be used to perform various downstreamtasks, such as clustering DNA sequences and labs, as well as using them as features in machine learning models. Our approach employsa circular shift augmentation approach and is able to correctly rank the lab-of-origin90%of the time within its top 10 predictions -outperforming all current state-of-the-art approaches. We also demonstrate that we can perform few-shot-learning and obtain76%top-10 accuracy using only10%of the sequences. This means, we outperform the previous CNN approach using only one-tenth of the data. We also demonstrate that we are able to extract key signatures in plasmid sequences for particular labs, allowing for an interpretable examination of the model’s outputs.CCS Concepts: Information systems→Similarity measures; Learning to rank.

Influence of Temperature and Backslopping Time on the Microbiota of a Type I Propagated Laboratory Wheat Sourdough Fermentation

ABSTRACTSourdough fermentation is a cereal fermentation that is characterized by the formation of stable yeast/lactic acid bacteria (LAB) associations. It is a unique process among food fermentations in that the LAB that mostly dominate these fermentations are heterofermentative. In the present study, four wheat sourdough fermentations were carried out under different conditions of temperature and backslopping time to determine their effect on the composition of the microbiota of the final sourdoughs. A substantial effect of temperature was observed. A fermentation with 10 backsloppings (once every 24 h) at 23°C resulted in a microbiota composed ofLeuconostoc citreumas the dominant species, whereas fermentations at 30 and 37°C with backslopping every 24 h resulted in ecosystems dominated byLactobacillus fermentum. Longer backslopping times (every 48 h at 30°C) resulted in a combination ofLactobacillus fermentumandLactobacillus plantarum. Residual maltose remained present in all fermentations, except those with longer backslopping times, and ornithine was found in almost all fermentations, indicating enhanced sourdough-typical LAB activity. The sourdough-typical speciesLactobacillus sanfranciscensiswas not found. Finally, a nonflour origin for this species was hypothesized.