Genome-wide copy number analysis using copy number inferring tool (CNIT) and DNA pooling.

Published

Journal Article

Copy number variation (CNV) has become an important genomic structure element in the human population, and some CNVs are related to specific traits and diseases. Moreover, analysis of human genomes has been potentiated by the use of high-resolution microarrays that assess single nucleotide polymorphisms (SNPs). Although many programs have been designed to analyze data from Affymetrix SNP microarrays, they all have high false-positive rates (FPRs) in copy number (CN) analyses. Copy number analysis tool (CNAT) 4.0 is a recently developed program that offers improved CN estimation, but small amplifications and deletions are lost when using the smoothing procedure. Here, we propose a copy number inferring tool (CNIT) algorithm for the 100K SNP microarray to investigate CNVs at 29.6-kb resolution. CNIT estimated SNP allelic and total CN with reliable P values based on intensity data. In addition, the hidden Markov model (HMM) method was applied to predict regions having altered CN by considering contiguous SNPs. Based on a CN analysis of 23 unrelated Taiwanese and 30 HapMap Centre d'Etude du Polymorphisme Humain (CEPH) trios, CNIT showed higher accuracy and power than other programs. The FPRs and false-negative rates (FNRs) of CNIT were 0.1% and 0.16%, respectively. CNIT also showed better sensitivity for detecting small amplifications and deletions. Furthermore, DNA pooling of 10 and 30 normal unrelated individuals were applied to the 100K SNP microarray, respectively, and 12 common CN-variable regions were identified, suggesting that DNA pooling can be applied to discover common CNVs.

Full Text

Duke Authors

Cited Authors

  • Lin, C-H; Huang, M-C; Li, L-H; Wu, J-Y; Chen, Y-T; Fann, CSJ

Published Date

  • August 2008

Published In

Volume / Issue

  • 29 / 8

Start / End Page

  • 1055 - 1062

PubMed ID

  • 18470944

Pubmed Central ID

  • 18470944

Electronic International Standard Serial Number (EISSN)

  • 1098-1004

Digital Object Identifier (DOI)

  • 10.1002/humu.20760

Language

  • eng

Conference Location

  • United States