Evidential deep learning for trustworthy prediction of enzyme commission number.

The rapid growth of uncharacterized enzymes and their functional diversity urge accurate and trustworthy computational functional annotation tools. However, current state-of-the-art models lack trustworthiness on the prediction of the multilabel classification problem with thousands of classes. Here, we demonstrate that a novel evidential deep learning model (named ECPICK) makes trustworthy predictions of enzyme commission (EC) numbers with data-driven domain-relevant evidence, which results in significantly enhanced predictive power and the capability to discover potential new motif sites. ECPICK learns complex sequential patterns of amino acids and their hierarchical structures from 20 million enzyme data. ECPICK identifies significant amino acids that contribute to the prediction without multiple sequence alignment. Our intensive assessment showed not only outstanding enhancement of predictive performance on the largest databases of Uniprot, Protein Data Bank (PDB) and Kyoto Encyclopedia of Genes and Genomes (KEGG), but also a capability to discover new motif sites in microorganisms. ECPICK is a reliable EC number prediction tool to identify protein functions of an increasing number of uncharacterized enzymes.

How 'who someone is' and 'what they did' influences gossiping about them.

To understand, predict, and help correct each other's actions we need to maintain accurate, up-to-date knowledge of people, and communication is a critical means by which we gather and disseminate this information. Yet the conditions under which we communication social information remain unclear. Testing hypotheses generated from our theoretical framework, we examined when and why social information is disseminated about an absent third party: i.e., gossiped. Gossip scenarios presented to participants (e.g., "Person-X cheated on their exam") were based on three key factors: (1) target (ingroup, outgroup, or celebrity), (2) valence (positive or negative), and (3) content. We then asked them (a) whether they would spread the information, and (b) to rate it according to subjective valence, ordinariness, interest level, and emotion. For ratings, the scenarios participants chose to gossip were considered to have higher valence (whether positive or negative), to be rarer, more interesting, and more emotionally evocative; thus showing that the paradigm was meaningful to subjects. Indeed, for target, valence, and content, a repeated-measures ANOVA found significant effects for each factor independently, as well as their interactions. The results supported our hypotheses: e.g., for target, more gossiping about celebrities and ingroup members (over strangers); for valence, more about negative events overall, and yet for ingroup members, more positive gossiping; for content, more about moral topics, with yet all domains of social content communicated depending on the situation-context matters, influencing needs. The findings suggest that social knowledge sharing (i.e., gossip) involves sophisticated calculations that require our highest sociocognitive abilities, and provide specific hypotheses for future examination of neural mechanisms.

Micro environment data on the Daeungbojeon hall and the Palsangjeon hall of the Beopjusa temple in Republic of Korea.

This article provides long-term environmental change data for wooden buildings; it also reflects environmental data provided by the Korea Meteorological Administration. In the case of field survey, data logger was installed on the left rear and right front sides of the buildings. Datasets on the Beopjusa temple were collected at 1 h intervals in each building. Korea Meteorological Administration data was collected from public database(data.kma.go.kr) and all data processed in excel. The data was collected at two sites from Daeungbojeon hall and Palsangjeon hall in the Beopjusa temple, Republic of Korea. Data sets at 1 h intervals are provided by collecting more than 170,000 pieces of data for each building. And monthly average dataset and difference value of time average data between inside and outside are provided. This data can be used as basic data for environmental change researcher or simulation researcher of wood condition.