Somatic structure variations (SVs) and copy number variations (CNVs) may result in genetic changes that are directly or indirectly related to different types of neoplasm. Computational tools have been developed to detect structural and copy number variations from next-generation sequencing (NGS) data. However, with no prior knowledge about variants in real samples, those tools have been hindered by the lack of a gold standard benchmark.
MeSHgram is a tool for convenient, visual and interactive exploration of the co-occurrence of MeSH terms over the entire PubMed corpus. Such a tool can assist in quantification of known research patterns as well as potentially aiding novel hypotheses generation.
A crowd-sourced platform for algorithms and analysis pipelines focused on time- and state-discrete dynamical systems. It features an easy way for developers to publish their own algorithms and link them with others to create workflows for the analysis and use of systems within mathematics and in applications to other fields such as biology and engineering.
There is a growing need in bioinformatics (and computational science in general) for easy-to-use software tools that facilitates building and troubleshooting pipelines. At the same time, available tools requires that users pass through a lengthy process of local installations and configurations for individual modules. Testing mathematical models using pipelines becomes a hurdle for researchers.
A Deep Learning Approach: In the past, the study of medicine used to focus on observing biological processes that take place in organisms, and based on these observations, biologists would make conclusions that translate into a better understanding of how organisms systems work. Recently, the approach has changed to a computational paradigm, where scientists try to model these biological processes as mathematical equations or statistical models. In this study, we have modeled an important activity of cell replication in a type of bacteria genome using different deep learning network models. Results from this research suggest that deep learning models have the potential to learn representations of DNA sequences, hence predicting cell behavior.