Background: High density single nucleotide polymorphism microarrays (SNP chips) provide information on a subject's genome, such as the chromosomal copy numbers and the genotype (heterozygosity/homozygosity). In contrast to fluorescence in situ hybridization and karyotyping, SNP chips provide a high resolution map of the human genome that can be used to detect, for example, microdeletions, microduplications, and loss of heterozygosity. As a variety of diseases are linked to such chromosomal alterations, SNP chips promise new insights for these diseases by aiding in the discovery of such regions, and may suggest targets for intervention. The R package VanillaICE contains the software for fitting hidden Markov models on genomic array data to infer chromosomal alterations, including deletions, amplifications, and regions with loss of heterozygosity. In addition, measures of uncertainty for the genotype and copy number estimates can be incorporated, which can be crucial for the detection of micro-deletions and micro-amplifications.

Collaborators: VanillaICE, developed in the Department of Biostatistics at the Johns Hopkins Bloomberg School of Public Health, is a collaboration with Rob Scharpf and Giovanni Parmigiani.

Code: The R package is available from the Bioconductor webpage as Free Software in source code form under the terms of the GNU General Public License of the Free Software Foundation. The accompanying Annals of Applied Statistics manuscript is available here, and supplementary material is here.