Format

Send to

Choose Destination
Sci Rep. 2015 May 15;5:10247. doi: 10.1038/srep10247.

Evaluating information content of SNPs for sample-tagging in re-sequencing projects.

Author information

1
Department of human molecular genetics, Max-Planck Institute for Molecular Genetics, Berlin, 14195, Germany.
2
BlackBerry Deutschland GmbH, Bochum, 44799, Germany.
3
Systems Biology Center, National Heart, Lung, and Blood Institute, NIH, Bethesda, MD, 20892, USA.

Abstract

Sample-tagging is designed for identification of accidental sample mix-up, which is a major issue in re-sequencing studies. In this work, we develop a model to measure the information content of SNPs, so that we can optimize a panel of SNPs that approach the maximal information for discrimination. The analysis shows that as low as 60 optimized SNPs can differentiate the individuals in a population as large as the present world, and only 30 optimized SNPs are in practice sufficient in labeling up to 100 thousand individuals. In the simulated populations of 100 thousand individuals, the average Hamming distances, generated by the optimized set of 30 SNPs are larger than 18, and the duality frequency, is lower than 1 in 10 thousand. This strategy of sample discrimination is proved robust in large sample size and different datasets. The optimized sets of SNPs are designed for Whole Exome Sequencing, and a program is provided for SNP selection, allowing for customized SNP numbers and interested genes. The sample-tagging plan based on this framework will improve re-sequencing projects in terms of reliability and cost-effectiveness.

PMID:
25975447
PMCID:
PMC4432563
DOI:
10.1038/srep10247
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Nature Publishing Group Icon for PubMed Central
Loading ...
Support Center