Introduction to biomartr

2018-01-02

Getting Started

A major problem in bioinformatics research is consistent data retrieval. The biomartr package aims to fill this gap by providing users with easy to use and diverse interfaces to well curated genomic databases such as:

The collection of functions implemented in biomartr enable fast functional annotation queries for a set of genes, entire genomes or meta-genome projects such as biological pathway analyses and genomic sequence retrieval. Although, the name of this package biomartr is inspired by the Ensembl Biomart database it extends its functionality and service by providing additional interface functions to the most prominent databases for genome, proteime, CDS, GFF, etc. retrieval.

Workflow Design

The biomartr package is designed to cover the following branches of biological data retrieval:

Hence, all functions implemented in biomartr serve to fulfill the above listed tasks.

Biological Sequence Retrieval

In the post-genomic era, biological sequences are used to investigate most phenomena of molecular biology. The growing number of databases and their entries allows us to design meta-studies in a new dimension and to re-investigate known phenomena from a new perspective. Neverthless, from a data science point of view this vast amount of heterogenous data, coming from very different data resources and data standards is very hard to transform from heterogenous to homogenous data. The detection of significant patterns within meta-analyses, therefore relies on high quality data analysis and data science.

Another aspect is reproducibility. Even in cases where a high degree of data homogeneity is achieved, the aspect of scientific reproducibility adds up to a new layer of complexity. Much effort is now being invested to enable high standards of reproducibility in data driven sciences (e.g. ROpenSci). The biomartr package aims to be a part of this data science movement. It’s functions implement interfaces (Application Programming Interfaces, short APIs) to major databases such as NCBI, ENSEMBL, and ENSEMBLGENOMES allowing users to access curated data from major genome data sources.

The Sequence Retrieval and Metagenome Retrieval vignettes will introduce users to the process of genomic sequence retrieval using biomartr. All functions were designed to allow users to achieve the highest (yet possible) degree of reproducibility and transparency for their own analyses.

Functional Annotation Retrieval

Most phenomena in molecular biology are based on the interaction of genes and their products (proteins). So in case patterns can be found by investigating biological sequences, in most cases it is of interest to map this pattern to a biological function. The Gene Ontology and Kegg consortia provide extensive information on gene functions and pathway memberships.

The biomartr packages provides powerful interfaces to functional annotation for specific genes. Here, users can consult the Functional Annotation vignette which introduces them to the process of functional information retrieval.