A protein-free chemically defined medium for the cultivation of various micro-organisms with food safety significance

Development of chemically defined media for Lactococcus lactis subsp. lactis YF11 to eliminate the influence of hyperosmotic stress

Chemically defined media (CDM) can eliminate or lessen the interference that occurs in complex culture media (CCM) caused by the undefined substrate pools, and various CDM have been designed and employed for investigating microbial physiology and multiomics. Herein, using the measured amount of total amino acids in CCM and combined with the in vivo and in vitro amino acid content of Lactococcus lactis subsp. lactis YF11, new enriched CDM were designed and then optimized using a statistical design-of-experiment method coupling with fed-batch fermentation to eliminate or lessen the influence of hyperosmotic pressure. Cell volume was introduced as a target index to assess the performance of CDM, and average osmotic pressure (AOP) was employed to describe the osmotic pressure of CDM. The AOP was significantly decreased from 610 mOsm/kg·H2O in the initial CDM (I-CDM) to 360 mOsm/kg·H2O in fed-batch CDM (F-CDM), and the cell volume was increased from 0.142 ± 0.004 μm3 in I-CDM to 0.198 ± 0.008 μm3 in F-CDM, which was close to 0.206 ± 0.005 μm3 found in CCM, indicating that the strategy of designing and improving CDM followed by a statistical design-of-experiment coupling with fed-batch cultivation presented a promising pathway for extensive utilization of CDM.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-023-03788-5.

Feature impact assessment: a new score to identify relevant metabolomics features in artificial neural networks using validated labels

INTRODUCTION

Artificial Neural Networks (ANN) are increasingly used in metabolomics but are hard to interpret.

OBJECTIVES

We aimed at developing a feature impact score that is model-agnostic, simple, and interpretable.

METHODS

Feature Impact Assessment (FIA) is calculated by varying combinations of features within their observed value range and checking for changes in prediction outcomes. FIA was implemented in R and tested on metabolomics datasets.

RESULTS

FIA exceeded LIME and SHAP in selecting biologically meaningful features. Values were comparable across different ANN architectures.

CONCLUSION

FIA is a novel score ranking feature impact, helping interpreting ANN in the metabolomics field.

Deep Learning for Rapid Identification of Microbes Using Metabolomics Profiles

Rapid detection of viable microbes remains a challenge in fields such as microbial food safety. We here present the application of deep learning algorithms to the rapid detection of pathogenic and non-pathogenic microbes using metabolomics data. Microbes were incubated for 4 h in a protein-free defined medium, followed by 1D ¹H nuclear magnetic resonance (NMR) spectroscopy measurements. NMR spectra were analyzed by spectral binning in an untargeted metabolomics approach. We trained multilayer ("deep") artificial neural networks (ANN) on the data and used the resulting models to predict spectra of unknown microbes. ANN predicted unknown microbes in this laboratory setting with an average accuracy of 99.2% when using a simple feature selection method. We also describe learning behavior of the employed ANN and the optimization strategies that worked well with these networks for our datasets. Performance was compared to other current data analysis methods, and ANN consistently scored higher than random forest models and support vector machines, highlighting the potential of deep learning in metabolomics data analysis.