Single-cell RNA-sequencing (scRNA-Seq) is widely used to reveal the heterogeneity and dynamics of tissues, organisms, and complex diseases, but its analyses still suffer from multiple grand challenges, including the sequencing sparsity and complex differential patterns in gene expression. We introduce the scGNN (single-cell graph neural network) to provide a hypothesis-free deep learning framework for scRNA-Seq analyses. This framework formulates and aggregates cell–cell relationships with graph neural networks and models heterogeneous gene expression patterns using a left-truncated mixture Gaussian model. scGNN integrates three iterative multi-modal autoencoders and outperforms existing tools for gene imputation and cell clustering on four benchmark scRNA-Seq datasets. In an Alzheimer’s disease study with 13,214 single nuclei from postmortem brain tissues, scGNN successfully illustrated disease-related neural development and the differential mechanism. scGNN provides an effective representation of gene expression and cell–cell relationships. It is also a powerful framework that can be applied to general scRNA-Seq analyses.

Circular RNAs (circRNAs) are a unique class of RNA molecule identified more than 40 years ago which are produced by a covalent linkage via back-splicing of linear RNA. Recent advances in sequencing technologies and bioinformatics tools have led directly to an ever-expanding field of types and biological functions of circRNAs. In parallel with technological developments, practical applications of circRNAs have arisen including their utilization as biomarkers of human disease. Currently, circRNA-associated bioinformatics tools can support projects including circRNA annotation, circRNA identification and network analysis of competing endogenous RNA (ceRNA). In this review, we collected about 100 circRNA-associated bioinformatics tools and summarized their current attributes and capabilities. We also performed network analysis and text mining on circRNA tool publications in order to reveal trends in their ongoing development.

BackgroundThe Internet has become one of the most important means to obtain health and medical information. It is often the first step in checking for basic information about a disease and its treatment. The search results are often useful to general users. Various search engines such as Google, Yahoo!, Bing, and Ask.com can play an important role in obtaining medical information for both medical professionals and lay people. However, the usability and effectiveness of various search engines for medical information have not been comprehensively compared and evaluated.ObjectiveTo compare major Internet search engines in their usability of obtaining medical and health information.MethodsWe applied usability testing as a software engineering technique and a standard industry practice to compare the four major search engines (Google, Yahoo!, Bing, and Ask.com) in obtaining health and medical information. For this purpose, we searched the keyword breast cancer in Google, Yahoo!, Bing, and Ask.com and saved the results of the top 200 links from each search engine. We combined nonredundant links from the four search engines and gave them to volunteer users in an alphabetical order. The volunteer users evaluated the websites and scored each website from 0 to 10 (lowest to highest) based on the usefulness of the content relevant to breast cancer. A medical expert identified six well-known websites related to breast cancer in advance as standards. We also used five keywords associated with breast cancer defined in the latest release of Systematized Nomenclature of Medicine-Clinical Terms (SNOMED CT) and analyzed their occurrence in the websites.ResultsEach search engine provided rich information related to breast cancer in the search results. All six standard websites were among the top 30 in search results of all four search engines. Google had the best search validity (in terms of whether a website could be opened), followed by Bing, Ask.com, and Yahoo!. The search results highly overlapped between the search engines, and the overlap between any two search engines was about half or more. On the other hand, each search engine emphasized various types of content differently. In terms of user satisfaction analysis, volunteer users scored Bing the highest for its usefulness, followed by Yahoo!, Google, and Ask.com.ConclusionsGoogle, Yahoo!, Bing, and Ask.com are by and large effective search engines for helping lay users get health and medical information. Nevertheless, the current ranking methods have some pitfalls and there is room for improvement to help users get more accurate and useful information. We suggest that search engine users explore multiple search engines to search different types of health information and medical knowledge for their own needs and get a professional consultation if necessary.

We have developed a generally adaptable, novel high-throughput Viral Chromosome Conformation Capture assay (V3C-seq) for use in trans that allows genome-wide identification of the direct interactions of a lytic virus genome with distinct regions of the cellular chromosome. Upon infection, we found that the parvovirus Minute Virus of Mice (MVM) genome initially associated with sites of cellular DNA damage that in mock-infected cells also exhibited DNA damage as cells progressed through S-phase. As infection proceeded, new DNA damage sites were induced, and virus subsequently also associated with these. Sites of association identified biochemically were confirmed microscopically and MVM could be targeted specifically to artificially induced sites of DNA damage. Thus, MVM established replication at cellular DNA damage sites, which provide replication and expression machinery, and as cellular DNA damage accrued, virus spread additionally to newly damaged sites to amplify infection. MVM-associated sites overlap significantly with previously identified topologically-associated domains (TADs).

Genomic selection uses single-nucleotide polymorphisms (SNPs) to predict quantitative phenotypes for enhancing traits in breeding populations and has been widely used to increase breeding efficiency for plants and animals. Existing statistical methods rely on a prior distribution assumption of imputed genotype effects, which may not fit experimental datasets. Emerging deep learning technology could serve as a powerful machine learning tool to predict quantitative phenotypes without imputation and also to discover potential associated genotype markers efficiently. We propose a deep-learning framework using convolutional neural networks (CNNs) to predict the quantitative traits from SNPs and also to investigate genotype contributions to the trait using saliency maps. The missing values of SNPs are treated as a new genotype for the input of the deep learning model. We tested our framework on both simulation data and experimental datasets of soybean. The results show that the deep learning model can bypass the imputation of missing values and achieve more accurate results for predicting quantitative phenotypes than currently available other well-known statistical methods. It can also effectively and efficiently identify significant markers of SNPs and SNP combinations associated in genome-wide association study.

Salinity is one of the most common abiotic stresses in agriculture production. Salt tolerance of rice (Oryza sativa) is an important trait controlled by various genes. The mechanism of rice salt tolerance, currently with limited understanding, is of great interest to molecular breeding in improving grain yield. In this study, a gene regulatory network of rice salt tolerance is constructed using a systems biology approach with a number of novel computational methods. We developed an improved volcano plot method in conjunction with a new machine-learning method for gene selection based on gene expression data and applied the method to choose genes related to salt tolerance in rice. The results were then assessed by quantitative trait loci (QTL), co-expression and regulatory binding motif analysis. The selected genes were constructed into a number of network modules based on predicted protein interactions including modules of phosphorylation activity, ubiquity activity, and several proteinase activities such as peroxidase, aspartic proteinase, glucosyltransferase, and flavonol synthase. All of these discovered modules are related to the salt tolerance mechanism of signal transduction, ion pump, abscisic acid mediation, reactive oxygen species scavenging and ion sequestration. We also predicted the three-dimensional structures of some crucial proteins related to the salt tolerance QTL for understanding the roles of these proteins in the network. Our computational study sheds some new light on the mechanism of salt tolerance and provides a systems biology pipeline for studying plant traits in general.

Background: Mutations in SLC6A1 have been associated mainly with myoclonic atonic epilepsy (MAE) and intellectual disability. We identified a novel missense mutation in a patient with Lennox-Gastaut syndrome (LGS) characterized by severe seizures and developmental delay. Methods: Exome Sequencing was performed in an epilepsy patient cohort. The impact of the mutation was evaluated by 3H γ-aminobutyric acid (GABA) uptake, structural modeling, live cell microscopy, cell surface biotinylation and a high-throughput assay flow cytometry in both neurons and non neuronal cells. Results: We discovered a heterozygous missense mutation (c700G to A [pG234S) in the SLC6A1 encoding GABA transporter 1 (GAT-1). Structural modeling suggests the mutation destabilizes the global protein conformation. With transient expression of enhanced yellow fluorescence protein (YFP) tagged rat GAT-1 cDNAs, we demonstrated that the mutant GAT-1(G234S) transporter had reduced total protein expression in both rat cortical neurons and HEK 293T cells. With a high-throughput flow cytometry assay and live cell surface biotinylation, we demonstrated that the mutant GAT-1(G234S) had reduced cell surface expression. 3H radioactive labeling GABA uptake assay in HeLa cells indicated a reduced function of the mutant GAT-1(G234S). Conclusions: This mutation caused instability of the mutant transporter protein, which resulted in reduced cell surface and total protein levels. The mutation also caused reduced GABA uptake in addition to reduced protein expression, leading to reduced GABA clearance, and altered GABAergic signaling in the brain. The impaired trafficking and reduced GABA uptake function may explain the epilepsy phenotype in the patient.

The autonomous parvovirus Minute Virus of Mice (MVM) localizes to cellular DNA damage sites to establish and sustain viral replication centers, which can be visualized by focal deposition of the essential MVM non-structural phosphoprotein NS1. How such foci are established remains unknown. Here, we show that NS1 localized to cellular sites of DNA damage independently of its ability to covalently bind the 5’ end of the viral genome, or its consensus DNA binding sequence. Many of these sites were identical to those occupied by virus during infection. However, localization of the MVM genome to DNA damage sites occurred only when wild-type NS1, but not its DNA-binding mutant was expressed. Additionally, wild-type NS1, but not its DNA binding mutant, could localize a heterologous DNA molecule containing the NS1 binding sequence to DNA damage sites. These findings suggest that NS1 may function as a bridging molecule, helping the MVM genome localize to cellular DNA damage sites to facilitate ongoing virus replication.