作者: Y. Y. Teo , M. Inouye , K. S. Small , R. Gwilliam , P. Deloukas
DOI: 10.1093/BIOINFORMATICS/BTM443
关键词: Affymetrix genechip 、 Algorithm 、 Stability (learning theory) 、 Metric (mathematics) 、 Genotyping 、 Executable 、 Biology 、 Genotype 、 Software 、 Training set
摘要: Motivation: Large-scale genotyping relies on the use of unsupervised automated calling algorithms to assign genotypes hybridization data. A number such have been recently established for Affymetrix GeneChip technology. Here, we present a fast and accurate genotype algorithm Illumina BeadArray platforms. As technology moves towards assaying millions genetic polymorphisms simultaneously, there is need an integrated easy-to-use software genotypes. Results: We introduced model-based which does not rely having prior training data or require computationally intensive procedures. The can from thousands individuals simultaneously pools information across multiple improve calling. method accommodate variations in intensities result dramatic shifts position clouds by identifying optimal coordinates initialize algorithm. By incorporating process perturbation analysis, obtain quality metric measuring stability assigned calls. show that this be used identify SNPs with low call rates accuracy. Availability: C++ executable described here available request authors. Contact:teo@well.ox.ac.uk tgc@well.ox.ac.uk