An embedding is a function that maps entities from one algebraic structure into another while preserving certain characteristics. Embeddings are being used successfully for mapping relational data or text into vector spaces where they can be used for machine learning, similarity search, or similar tasks. We address the problem of finding vector space embeddings for theories in the Description Logic EL ++ that are also models of the TBox. To find such embeddings, we define an optimization problem that characterizes the model-theoretic semantics of the operators in EL ++ within R n , thereby solving the problem of finding an interpretation function for an EL ++ theory given a particular domain ∆. Our approach is mainly relevant to large EL ++ theories and knowledge bases such as the ontologies and knowledge graphs used in the life sciences. We demonstrate that our method can be used for improved prediction of protein-protein interactions when compared to semantic similarity measures or knowledge graph embeddings.

Motivation Infectious diseases caused by novel viruses have become a major public health concern. Rapid identification of virus–host interactions can reveal mechanistic insights into infectious diseases and shed light on potential treatments. Current computational prediction methods for novel viruses are based mainly on protein sequences. However, it is not clear to what extent other important features, such as the symptoms caused by the viruses, could contribute to a predictor. Disease phenotypes (i.e., signs and symptoms) are readily accessible from clinical diagnosis and we hypothesize that they may act as a potential proxy and an additional source of information for the underlying molecular interactions between the pathogens and hosts. Results We developed DeepViral, a deep learning based method that predicts protein–protein interactions (PPI) between humans and viruses. Motivated by the potential utility of infectious disease phenotypes, we first embedded human proteins and viruses in a shared space using their associated phenotypes and functions, supported by formalized background knowledge from biomedical ontologies. By jointly learning from protein sequences and phenotype features, DeepViral significantly improves over existing sequence-based methods for intra- and inter-species PPI prediction. Availability Code and datasets for reproduction and customization are available at https://github.com/bio-ontology-research-group/DeepViral. Prediction results for 14 virus families are available at https://doi.org/10.5281/zenodo.4429824.

Background/Aims: Previous studies have suggested that myricetin (Myr) could promote the expression and nuclear translocation of nuclear factor (erythroid-derived 2)-like (Nrf2). This study aimed to investigate whether Myr could attenuate diabetes-associated kidney injuries and dysfunction in wild-type (WT) and Nrf2 knockdown (Nrf2-KD) mice. Methods: Lentivirus-mediated Nrf2-KD and WT mice were used to establish type 1 diabetes mellitus (DM) by streptozotocin (STZ) injection. WT and Nrf2-KD mice were then randomly allocated into four groups: control (CON), Myr, STZ, and STZ + Myr. Myr (100 mg/kg/day) or vehicle was administered for 6 months. Kidneys were harvested and weighed at the end of the experiment. Hematoxylin and eosin staining and Masson’s trichrome staining were used to assess the morphology and fibrosis of the kidneys, respectively. Urinary albumin-to-creatinine ratio was used to test renal function. Western blotting was performed to determine oxidative-stress- or inflammation-associated signaling pathways. Real-time polymerase chain reaction (RT-PCR) was performed to detect the expression of fibrosis or inflammatory cytokines at the message Ribonucleic Acid (mRNA) level. Results: In WT mice, Myr alleviated DM-induced renal dysfunction, fibrosis, and oxidative damage and enhanced the expression of Nrf2 and its downstream genes. After knockdown of Nrf2, Myr treatment partially but significantly mitigated DM-induced renal dysfunction and fibrosis, which might be associated with inhibition of the I-kappa-B (IκB)/nuclear factor-κB (NF-κB) (P65) signaling pathway. Conclusions: This study showed that Myr prevented DM-associated decreased expression of Nrf2 and inhibited IκB/NF-κB (P65) signaling pathway. Moreover, inhibition of IκB/NF-κB (P65) signaling pathway is independent of the regulation of Nrf2. Thus, Myr could be a potential treatment for preventing the development and progression of DM-associated kidney injuries and dysfunction.

No abstract

Motivation: Direct RNA sequencing (dRNA-seq) on the Oxford Nanopore Technology platforms has become increasingly popular in recent years, with promising outlook to transform the field of epitranscriptomics. Reads produced from dRNA-seq can cover up to full-length gene transcripts while containing decipherable information about RNA base modifications and poly-A tail lengths. Although many studies have been published exploring and expanding the potential of dRNA-seq, the sequencing accuracy and error patterns remain understudied and less characterized compared to DNA sequencing. Results: We evaluated the sequencing accuracy and characterized the systematic error patterns of dRNA-seq on public datasets including native RNA samples from diverse organisms, as well as synthetic in vitro transcribed RNAs. The median read accuracy is about 90% for most organisms, although some species are more challenging. Deletions account for the majority of errors and are twice as common as mismatches or insertions. Apart from the well-known homopolymer sequencing errors, there are systematic biases across all organisms at both single nucleotide and 2-mer motif level. In general, cytosines and uracils are more likely to be erroneous than guanines and adenines. Moreover, the systematic errors are found to be strongly dependent on the local sequence contexts, with complex interactions between adjacent positions. We further evaluated the accuracy of sequencing homopolymers, read quality scores as an estimate of error rates, and the consequences of failing to detect the DNA adaptors. Lastly, we discuss the relevance of such error patterns for the downstream applications of dRNA-seq data, such as transcript identification and base modification detection.

Motivation: Infectious diseases from novel viruses are becoming a major public health concern. Fast identification of virus-host interactions can reveal mechanistic insights of infectious diseases and shed light on potential treatments and drug discoveries. Current computational prediction methods for novel viruses are based only on protein sequences. Yet, it is not clear to what extent other important features, such as the symptoms caused by the viruses, could contribute to a predictor. Disease phenotypes (i.e., symptoms) are readily accessible from clinical diagnosis and we hypothesize that they may act as a potential proxy and an additional source of information for the underlying molecular interactions between the pathogens and hosts. Results: We developed DeepViral, a deep learning method that predicts potential protein-protein interactions between human and viruses. First, human proteins and viruses were embedded in a shared space using their associated phenotypes, functions, taxonomic classification, as well as formalized background knowledge from biomedical ontologies. By extending a sequence learning model with phenotype features, our model can not only significantly improve over previous sequence-based approaches for inter-species interaction prediction, but also identify pathways of viral targets under a realistic experimental setup for novel viruses. Availability:https://github.com/bio-ontology-research-group/DeepViral Contact:

Contemporary metagenomic annotation methods have proven insufficient in our attempts to better understand the complex environments around us. We call the yet to be annotated part of a metagenome it’s ‘dark matter’. The Gene Ontology (GO) is a hierarchical vocabulary used to describe gene product function and a large collection of curated genes with GO annotations already exists. DeepGO utilises deep learning to build models from these curated genes and gene products to predict GO categories for novel proteins. One of the major problems with metagenomic studies today is the process of assembling the environmental DNA sequences into their original genomes. This is difficult, with chimeric metagenomically assembled genomes being common. To avoid this and the computational and time expense, we have modified DeepGO to perform protein function prediction directly from sequence reads with limited protein coding sequence prediction. Three independent models were trained as the following; The first 50 amino acids of a protein were used for training, The last 50 amino acids were used for training, A phasing window of 50 amino acids was used to train across the entirety of a protein sequence. These models were chosen to learn from the different parts of a protein sequence we are likely to capture from only the short unassembled sequence reads. We compared the three models by producing a mock metagenomic community consisting of 6 model bacterial genomes. We evaluated the functions predicted from the unassembled sequence reads and the protein coding sequences predicted from the assembled metagenome.

Motivation: Drug-drug interactions (DDIs) are complex processes which may depend on many clinical and non-clinical factors. Identifying and distinguishing ways in which drugs interact remains a challenge. To minimize DDIs and to personalize treatment based on accurate stratification of patients, it is crucial that mechanisms of interaction can be identified. Most DDIs are a consequence of metabolic mechanisms of interaction, but DDIs with different mechanisms occur less frequently and are therefore more difficult to identify. Results: We developed a method (D4) for computationally identifying potential DDIs and determining whether they interact based on one of eleven mechanisms of interaction. D4 predicts DDIs and their mechanisms through features that are generated through a deep learning approach from phenotypic and functional knowledge about drugs, their side-effects and targets. Our findings indicate that our method is able to identify known DDIs with high accuracy and that D4 can determine mechanisms of interaction. We also identify numerous novel and potential DDIs for each mechanism of interaction and evaluate our predictions using DDIs from adverse event reporting systems.