The increase in microbial sequenced genomes from pure cultures and metagenomic samples reflects the current attainability of whole-genome and shotgun sequencing methods. However, software for genome visualization still lacks automation, integration of different analyses, and customizable options for non-experienced users. In this study, we introduce GenoVi, a Python command-line tool able to create custom circular genome representations for the analysis and visualization of microbial genomes and sequence elements. It is designed to work with complete or draft genomes, featuring customizable options including 25 different built-in color palettes (including 5 color-blind safe palettes), text formatting options, and automatic scaling for complete genomes or sequence elements with more than one replicon/sequence. Using a Genbank format file as the input file or multiple files within a directory, GenoVi (i) visualizes genomic features from the GenBank annotation file, (ii) integrates a Cluster of Orthologs Group (COG) categories analysis using DeepNOG, (iii) automatically scales the visualization of each replicon of complete genomes or multiple sequence elements, (iv) and generates COG histograms, COG frequency heatmaps and output tables including general stats of each replicon or contig processed. GenoVi's potential was assessed by analyzing single and multiple genomes of Bacteria and Archaea. Paraburkholderia genomes were analyzed to obtain a fast classification of replicons in large multipartite genomes. GenoVi works as an easy-to-use command-line tool and provides customizable options to automatically generate genomic maps for scientific publications, educational resources, and outreach activities. GenoVi is freely available and can be downloaded from https://github.com/robotoD/GenoVi.

An increasing amount of researchers use software images to capture the requirements and code dependencies needed to carry out computational experiments. Software images preserve the computational environment required to execute a scientific experiment and have become a crucial asset for reproducibility. However, software images are usually not properly documented and described, making it challenging for scientists to find, reuse and understand them. In this paper, we propose a framework for automatically describing software images in a machine-readable manner by (i) creating a vocabulary to describe software images; (ii) developing an annotation framework designed to automatically document the underlying environment of software images and (iii) creating DockerPedia, a Knowledge Graph with over 150,000 annotated software images, automatically described using our framework. We illustrate the usefulness of our approach in finding images with specific software dependencies, comparing similar software images, addressing versioning problems when running computational experiments; and flagging problems with vulnerable software dependencies.

The SPARQL query language is the standard for querying RDF data and has been implemented in a wide variety of engines. These engines support hundreds of public endpoints on the Web which receive thousands of queries daily. In many cases these endpoints struggle when evaluating complex queries or when they receive too many of them concurrently. They struggle mostly since some of these queries need large amounts of resources to be processed. All these engines have an internal query optimizer that proposes a supposedly optimal query execution plan, however this is a hard task since there may be thousands of possible query plans to consider and the optimizer may not chose the best one. Herein we propose the use of machine learning techniques to help in finding the best query plan for a given query fast, and thus improve the SPARQL servers' performance. We base such optimization in modeling SPARQL queries based on their complexity, operators used within the queries and data accessed, among others. In this work we propose the use of Dense Neural Networks to improve such SPARQL query processing times. Herein we present the general architecture of a neural network for optimizing SPARQL queries and the results over a synthetic benchmark and real world queries. We show that the use of Dense Neural Networks improve the performance of the Nu-SVR approach in about 50% in performance. We also contribute to the community with a dataset of 19,000 queries.

As the adoption of knowledge graphs grows, more and more non-experts users need to be able to explore and query such graphs. These users are not typically familiar with graph query languages such as SPARQL, and may not be familiar with the knowledge graph's structure. In this extended abstract, we provide a summary of our work on a language and visual interface -- called RDF Explorer -- that help non-expert users to navigate and query knowledge graphs. A usability study over Wikidata shows that users successfully complete more tasks with RDF Explorer than with the existing Wikidata Query Helper interface.

As more and more companies are augmenting their data to include semantics it is imperative that the choices made when choosing the modelling language are well founded in knowledge about the language and the domain in question. This work demonstrates how the semiotic quality framework can facilitate the choice of the most suited language for a real world application. Computational and situated features are introduced as an extension to the framework.