BIONIC: biological network integration using convolutions

Biological networks constructed from varied data can be used to map cellular function, but each data type has limitations. Network integration promises to address these limitations by combining and automatically weighting input information to obtain a more accurate and comprehensive representation of the underlying biology. We developed a deep learning-based network integration algorithm that incorporates a graph convolutional network framework. Our method, BIONIC (Biological Network Integration using Convolutions), learns features that contain substantially more functional information compared to existing approaches. BIONIC has unsupervised and semisupervised learning modes, making use of available gene function annotations. BIONIC is scalable in both size and quantity of the input networks, making it feasible to integrate numerous networks on the scale of the human genome. To demonstrate the use of BIONIC in identifying new biology, we predicted and experimentally validated essential gene chemical-genetic interactions from nonessential gene profiles in yeast.

Mutations of Kluyveromyces lactis dolichol kinase enhances secretion of recombinant proteins

Although there are similarities in the core steps of the secretion pathway from yeast to higher eukaryotes, significant functional differences exist even among diverse yeast species. Here, we used next-generation sequencing to identify two mutations in the Kluyveromyces lactis KlSEC59 gene, encoding dolichol kinase (DK), which are responsible for an enhanced secretion phenotype in a previously isolated mutant, MD2/1-9. Compared with the temperature-sensitive Saccharomyces cerevisiae sec59-1 mutant, which exhibits reduced N-glycosylation and decreased secretory efficacy, the identified K. lactis DK mutations had fewer effects on glycosylation, as well as on survival at high temperature and cell wall integrity. Moreover, despite some glycosylation defects, double DK mutations (G405S and I419S) in the K. lactis mutant strain demonstrated three times the level of recombinant α-amylase secretion as the wild-type strain. Overexpression of potential suppressors KlMNN10, KlSEL1, KlERG20, KlSRT1, KlRER2, KlCAX4, KlLPP1 and KlDPP1 in the DK-mutant strain restored carboxypeptidase Y glycosylation to different extents and, with the exception of KISRT1, reduced α-amylase secretion to levels observed in wild-type cells. Our results suggest that enhanced secretion related to reduced activity of mutant DK in K. lactis results from mild glycosylation changes that affect activity of other proteins in the secretory pathway.

Proteins involved in flor yeast carbon metabolism under biofilm formation conditions

A lack of sugars during the production of biologically aged wines after fermentation of grape must causes flor yeasts to metabolize other carbon molecules formed during fermentation (ethanol and glycerol, mainly). In this work, a proteome analysis involving OFFGEL fractionation prior to LC/MS detection was used to elucidate the carbon metabolism of a flor yeast strain under biofilm formation conditions (BFC). The results were compared with those obtained under non-biofilm formation conditions (NBFC). Proteins associated to processes such as non-fermentable carbon uptake, the glyoxylate and TCA cycles, cellular respiration and inositol metabolism were detected at higher concentrations under BFC than under the reference conditions (NBFC). This study constitutes the first attempt at identifying the flor yeast proteins responsible for the peculiar sensory profile of biologically aged wines. A better metabolic knowledge of flor yeasts might facilitate the development of effective strategies for improved production of these special wines.

Protein O-mannosylation: what we have learned from baker's yeast

BACKGROUND

Protein O-mannosylation is a vital type of glycosylation that is conserved among fungi, animals, and humans. It is initiated in the endoplasmic reticulum (ER) where the synthesis of the mannosyl donor substrate and the mannosyltransfer to proteins take place. O-mannosylation defects interfere with cell wall integrity and ER homeostasis in yeast, and define a pathomechanism of severe neuromuscular diseases in humans.

SCOPE OF REVIEW

On the molecular level, the O-mannosylation pathway and the function of O-mannosyl glycans have been characterized best in the eukaryotic model yeast Saccharomyces cerevisiae. In this review we summarize general features of protein O-mannosylation, including biosynthesis of the mannosyl donor, characteristics of acceptor substrates, and the protein O-mannosyltransferase machinery in the yeast ER. Further, we discuss the role of O-mannosyl glycans and address the question why protein O-mannosylation is essential for viability of yeast cells.

GENERAL SIGNIFICANCE

Understanding of the molecular mechanisms of protein O-mannosylation in yeast could lead to the development of novel antifungal drugs. In addition, transfer of the knowledge from yeast to mammals could help to develop diagnostic and therapeutic approaches in the frame of neuromuscular diseases. This article is part of a Special Issue entitled: Functional and structural diversity of endoplasmic reticulum.