Comprehensive identification of pathogenic microbes and antimicrobial resistance genes in food products using nanopore sequencing-based metagenomics.

Foodborne pathogens, particularly antimicrobial-resistant (AMR) bacteria, remain a significant threat to global health. Given the limitations of conventional culture-based approaches, which are limited in scope and time-consuming, metagenomic sequencing of food products emerges as a promising solution. This method provides a fast and comprehensive way to detect the presence of pathogenic microbes and antimicrobial resistance genes (ARGs). Notably, nanopore long-read sequencing provides more accurate bacterial taxonomic classification in comparison to short-read sequencing. Here, we revealed the impact of food types and attributes (origin, retail place, and food processing methods) on microbial communities and the AMR profile using nanopore metagenomic sequencing. We analyzed a total of 260 food products, including raw meat, sashimi, and ready-to-eat (RTE) vegetables. Clostridium botulinum, Acinetobacter baumannii, and Vibrio parahaemolyticus were identified as the top three foodborne pathogens in raw meat and sashimi. Importantly, even with low pathogen abundance, higher percentages of samples containing carbapenem and cephalosporin resistance genes were identified in chicken and RTE vegetables, respectively. In parallel, our results demonstrated that fresh, peeled, and minced foods exhibited higher levels of pathogenic bacteria. In conclusion, this comprehensive study offers invaluable data that can contribute to food safety assessments and serve as a basis for quality indicators.

Natural statistics support a rational account of confidence biases.

Previous work has sought to understand decision confidence as a prediction of the probability that a decision will be correct, leading to debate over whether these predictions are optimal, and whether they rely on the same decision variable as decisions themselves. This work has generally relied on idealized, low-dimensional models, necessitating strong assumptions about the representations over which confidence is computed. To address this, we used deep neural networks to develop a model of decision confidence that operates directly over high-dimensional, naturalistic stimuli. The model accounts for a number of puzzling dissociations between decisions and confidence, reveals a rational explanation of these dissociations in terms of optimization for the statistics of sensory inputs, and makes the surprising prediction that, despite these dissociations, decisions and confidence depend on a common decision variable.

Between-subject correlation of heart rate variability predicts movie preferences.

We introduce a novel and simple method for assessing audiences' emotional responses to audiovisuals (e.g. films). Viewers (N = 21) watched movies and TV commercials from different genres while wearing photoplethysmography (PPG) optic sensors on their wrists. Heart rate variability (HRV) synchrony was observed among the audience. Based on this between-subject synchrony measure, we identified emotionally arousing segments from the materials. New participants (N = 24; N = 16) were then invited to watch these identified segments along with some randomly selected segments as control; they reported that the former was more engaging (effect size w = .67; w = .5). This finding was confirmed in an online study with a larger cohort (N = 300). While some specific effects varied depending on movie genre or gender, HRV-based editing generally performed better than the control. These findings suggest that HRV synchrony can be used as a new tool for audience psychology, and potentially also for automatically creating short trailers out of movies in a principled manner while taking into account the human perspective.