paper.md
In PeteHaitch/GenomicTuples: Representation and Manipulation of Genomic Tuples

title: 'Representation and Manipulation of Genomic Tuples in R' tags: - bioinformatics - computational biology authors: - name: Peter F Hickey orcid: 0000-0002-8153-6258 affiliation: Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health date: 17 May 2016 bibliography: paper.bib

Summary

GenomicTuples is an R/Bioconductor package [@R;@Bioconductor] that defines general purpose containers for storing and manipulating genomic tuples. A genomic tuple of size m is of the form chromosome:strand:{pos_1, pos_2, ..., pos_m} where pos_1 < pos_2 < ... < pos_m are positions along the chromosome. The difference between a genomic tuple and a genomic range/interval is like that of a difference between an ordered set and an interval. For example, the genomic 2-tuple chr3:+:{65, 77} differs from the genomic range chr3:+:[65, 77] by not including any of the intervening loci, chr3:+:66 to chr3:+:76.

GenomicTuples aims to provide functionality for manipulating tuples of genomic co-ordinates that are analogous to those available for genomic ranges in the popular GenomicRanges R/Bioconductor package [@GenomicRanges]. To that end, the GenomicTuples API mimics that of GenomicRanges. By extending classes defined in the GenomicRanges package, objects from the GenomicTuples package may be used as drop-in replacements for objects from the GenomicRanges package. This ensures easy interoperability with other popular Bioconductor packages, such as SummarizedExperiment [@SummarizedExperiment], and the availability of common operations, such as finding overlaps between genomic tuples and genomic features of interest.