News from the Bioconductor Project

The ‘News from the Bioconductor Project’ article from the 2014-1 issue.

The Bioconductor project provides tools for the analysis and comprehension of high-throughput genomic data. The 824 software packages available in Bioconductor can be viewed at http://bioconductor.org/packages/release/. Navigate packages using ‘biocViews’ terms and title search. Each package has an html page with a description, links to vignettes, reference manuals, and usage statistics. Start using Bioconductor and R version 3.1.0 with

  source("http://bioconductor.org/biocLite.R")
  biocLite()

Install additional packages and dependencies, e.g., deepSNV, with

  source("http://bioconductor.org/biocLite.R")
  biocLite("deepSNV")

Upgrade installed packages with

  source("http://bioconductor.org/biocLite.R")
  biocLite()

1 Bioconductor 2.14 Release Highlights

Bioconductor 2.14 was released on 14 April 2014. It is compatible with R 3.1.0 and consists of 824 software packages, 200 experiment data packages, and more than 860 current annotation packages. The release includes 77 new software packages and many updates and improvements to existing packages. The release announcement includes descriptions of new packages and updated NEWS files provided by current package maintainers.

New packages continue to represent a wide variety of research areas. Rariant identifies single nucleotide variants (SNVs) based on the difference of binomially distributed mismatch rates between matched samples. SomaticSignatures identifies the mutational signatures of SNVs. Filtering of SNVs based on inheritance models, amino acid change consequence, and minor allele frequency is offered through VariantFiltering. The monocle package, currently in the ‘devel’ branch, preforms differential expression and time series analysis for single-cell expression experiments. CRISPR/Cas is a compelling new molecular biology technique used for gene editing; CRIPSRseek helps find potential guide RNAs for input target sequences. CCREPE assesses the significance of similarity measures in ‘compositional’ data sets, as found in microbial abundance studies. MIMOSA models count data using Dirichlet-multinomial and beta-binomial mixtures with applications to single-cell assays. Machine learning methods such as SVM, Random Forest and CART are applied to RNASeq data in MLSeq; clustering and classification methods are used to summarize active paths in genome-scale metabolic and reaction networks in NetPathMiner. Bioconductor hosts a number of packages relevant to chemical compound discovery. The latest additions integrate bioinformatics and chemoinformatics into a molecular informatics platform in Rcpi, and performs alternating least squares analysis on chemical data in alsace.

In addition to these contributed packages, ‘Ranges’ infrastructure packages such as GenomicRanges, GenomicAlignments, and GenomicFeatures provide an extensive, mature and extensible framework for interacting with high throughput sequence data. One recent addition is the GenomeInfoDb package, which contains functions that allow translation between different chromosome sequence naming conventions (e.g., UCSC versus NCBI). Many packages rely on the Ranges infrastructure for interoperable, re-usable analysis; (Lawrence 2013) provides an introduction.

Our large collection of microarray, transcriptome and organism-specific annotation packages have been updated to include current information. Most of these packages now provide access to the ‘select’ interface (keys, columns, keytypes and select) which enable programmatic access to the databases they contain. The AnnotationHub, with 10,780 entries, complements our traditional offerings with diverse whole genome annotations from Ensembl, ENCODE, dbSNP, UCSC, and elsewhere.

2 Other activities

The Bioconductor Git-SVN Bridge allows developers to synchronize a GitHub repository with the canonical Bioconductor SVN package repository. Commits made in SVN are propagated to GitHub and vice versa. This was driven by developer requests for access to social coding features, such as issue tracking and pull requests. The service has been well received, with 73 bridges established as of June 2014.

The Bioconductor Amazon Machine Instance is now compatible with the StarCluster toolkit. This enhancement makes it straightforward to configure a cluster with nodes that communicate via MPI, SSH or Sun Grid Engine, and to control jobs via the BiocParallel and BatchJobs packages. Details are available at the AMI page http://www.bioconductor.org/help/bioconductor-cloud-ami/#using_cluster.

New Bioconductor package contributors are encouraged to consult the Package Guidelines and Package Submission sections of the Bioconductor web site, and use the new BiocCheck package, in addition to R CMD check, for guidance on conforming to Bioconductor package standards.

The Bioconductor web site advertises training and community events; mailing lists connect users with each other, to domain experts, and to maintainers eager to ensure that their packages satisfy the needs of leading edge approaches. Keep abreast of packages added to the ‘devel’ branch and other activities by following @Bioconductor on Twitter.

Bioconductor packages used

deepSNV, Rariant, SomaticSignatures, VariantFiltering, CRIPSRseek, CCREPE, MIMOSA, MLSeq, NetPathMiner, Rcpi, alsace, GenomicRanges, GenomicAlignments, GenomicFeatures, GenomeInfoDb, AnnotationHub, BiocParallel, BatchJobs, BiocCheck

Note

This article is converted from a Legacy LaTeX article using the texor package. The pdf version is the official version. To report a problem with the html, refer to CONTRIBUTE on the R Journal homepage.