Inductive inference of novel protein-molecule interactions using Heterogeneous Graph Transformer (HGT) AutoEncoder

Protein-molecule interactions are promoted by the physicochemical characteristics of the actors involved, but structural information alone does not capture expression patterns, localization and pharmacokinetics. In this work we propose an integrative strategy for protein-molecule interaction discovery that combines different layers of information through the use of convolutional operators on graph, and frame the problem as missing link prediction task on an heterogeneous graph constituted by three node types: 1) molecules 2) proteins 3) diseases. Physicochemical information of the actors are encoded using shallow embedding techniques (SeqVec, Mol2Vec, Doc2Vec respectively) and are supplied as feature vectors to a Graph AutoEncoer (GAE) that uses a Heterogeneous Graph Transformer (HGT) in the encoder module. We show in this work that HGT Autoencoder can be used to accurately recapitulate the protein-molecule interactions set and propose novel relationships in inductive settings that are grounded in biological and functional information extracted from the graph.

Gender Representation in the Lexical Functional Grammar

Gender marking is a language universal, although some languages have a stronger Gender-marking grammar. The Lexical Functional Grammar (LFG), a linguistic theory, has a set of rules and levels to render for Gender marking. Bornee and developed within the larger framework of the Generative Grammar, the Lexical Functional Grammar has become a standalone autonomous theoretical theory. This paper draws data from French language to present a comprehensive development of Gender-marking analysis within the Lexical Functional Grammar Framework. Fundamentally, the LSG posits for four phrase structures, which are the C-structure representing lexical entries, the F-structure, which deals with the functional information, the A-structure, which structures predicate-argument relationships, and the ơ-structure, which handles semantic representations. Although the grammatical gender is arrayed all over the four structures, it is mainly presented in this paper as a feature in the lexicon, typically integrated in the C-structure and F-structure mapping.

CyanoOmicsDB: an integrated omics database for functional genomic analysis of cyanobacteria

With their photosynthetic ability and established genetic modification systems, cyanobacteria are essential for fundamental and biotechnological research. Till now, hundreds of cyanobacterial genomes have been sequenced, and transcriptomic analysis has been frequently applied in the functional genomics of cyanobacteria. However, the massive omics data have not been extensively mined and integrated. Here, we describe CyanoOmicsDB (http://www.cyanoomics.cn/), a database aiming to provide comprehensive functional information for each cyanobacterial gene. CyanoOmicsDB consists of 8 335 261 entries of cyanobacterial genes from 928 genomes. It provides multiple gene identifiers, visualized genomic location, and DNA sequences for each gene entry. For protein-encoding genes, CyanoOmicsDB can provide predicted gene function, amino acid sequences, homologs, protein-domain super-families, and accession numbers for various public protein function databases. CyanoOmicsDB integrates both transcriptional and translational profiles of Synechocystis sp. PCC 6803 under various environmental culture coditions and genetic backgrounds. Moreover, CyanoOmicsDB includes 23 689 gene transcriptional start sites, 94 644 identified peptides, and 16 778 post-translation modification sites obtained from transcriptomes or proteomes of several model cyanobacteria. Compared with other existing cyanobacterial databases, CyanoOmicsDB comprises more datasets and more comprehensive functional information. CyanoOmicsDB will provide researchers in this field with a convenient way to retrieve functional information on cyanobacterial genes.

The Role of Information in Knowledge-How

Knowledge-how is the kind of knowledge implicated in skill employment and acquisition. Intellectualists claim that knowledge-how is a special type of propositional knowledge. Anti-intellectualists claim that knowledge-how is not propositional. We argue that both views face two open challenges. The first challenge pertains to the relationship between informational states and motor variability. The second pertains to the epistemic function of practice that leads to skill (and knowledge-how). The aim of this paper is to suggest a general conceptual framework based on functional information with both intellectualist and anti-intellectualist features. Our proposal, we argue, avoids the above challenges, and can further the debate on knowledge-how and skill.

Marketing Clues on the Label Raise the Purchase Intention of Genetically Modified Food

As more and more genetically modified foods (GMFs) must be labeled, adding more information to increase the willingness to buy genetically modified food has become the focus of scholars and enterprises. Most current studies have confirmed that the consumer attitudes and purchase intention toward GMFs are not good. This study aims to match consumers’ different information-processing mechanisms by adding marketing information clues and regulating their purchase intentions by contradictory attitudes towards GMFs. According to the interest demands of GMFs, the marketing clue information was divided into functional information and environmental information. Through two studies, we find that consumers are more inclined to environmental information than heuristic. Functional information is more attractive to males, and the young generation prefers ecological information. Consumers with high ambivalence towards genetically modified foods are more inclined to choose environmental attribute information.

Methoden für die durchgängige Anwendung einer EBOM mithilfe klassenbasierter Substitutionsobjekte

Innerhalb der Initiative zur vierten industriellen Revolution, bei der Menschen, Produkt und Maschine in Echtzeit vernetzt sind, geht der Trend weg von der Serienfertigung und hin zur „Losgröße 1“. Die ansteigende Anzahl an Varianten bringt jedoch auch höhere Verwaltungsaufwände mit sich. Um dem entgegenzuwirken wird die Methode der funktional substituierbaren Bauteile eingeführt. Diese ermöglicht die Verwaltung von technisch substituierbaren Komponenten innerhalb der Stücklisten einer konkreten Variante. Durch die Integration der Methoden FIM (Functional-Information-Model) und FIA (Functional-Information-Assembly) wird ein weiterer Schritt zur 3D-Modell-basierten Stückliste gemacht. Unter Zuhilfenahme dieser Methoden wird die notwendige Veränderung Darstellung der Struktur der im Produkt enthaltenen Bauteile während des Produktentstehungsprozesses beschrieben. Dabei werden die verwendeten Begriffe klar definiert und voneinander abgegrenzt.