Published online 19 November 2004
Nucleic Acids Research, Vol. 32 No. 20 © Oxford University Press 2004; all rights reserved
Novel algorithm for automated genotyping of microsatellites
1 Japan Biological Information Research Center, Japan Biological Informatics Consortium, Tokyo, Japan, 2 Hitachi Software Engineering Co., Ltd, Tokyo, Japan, 3 Department of Molecular Life Science, Course of Basic Medical Science and Molecular Medicine, Tokai University School of Medicine, Kanagawa, Japan, 4 Biological Information Research Center, National Institute of Advanced Industrial Science and Technology, Tokyo, Japan and 5 Center for Information Biology and DNA Data Bank of Japan, National Institute of Genetics, Shizuoka, Japan
* To whom correspondence should be addressed. Tel: +81 3 5780 2111; Fax: +81 3 5780 2049; Email: tmatsumoto{at}hitachisoft.jp
Received August 23, 2004; Revised and Accepted October 29, 2004
Microsatellites or short tandem repeats (STRs) are abundant in the human genome with easily assayed polymorphisms, providing powerful genetic tools for mapping both Mendelian and complex traits. Microsatellite genotyping requires detection of the products of polymerase chain reaction (PCR) amplification by electrophoresis, and analysis of the peak data for discrimination of the true allele. A high-throughput genotyping approach requires computer-based automation at both the detection and analysis phases. In order to achieve this, complicated peak patterns from individual alleles must be interpreted in order to assign alleles. Previous methods consider limited types of noise peaks and cannot provide enough accuracy. By pattern recognition of various types of noise peaks, such as stutter peaks and additional peaks, we have achieved an overall average accuracy of 94% for allele calling in our actual data. Our algorithm is crucial for a high-throughput genotyping system for microsatellite markers by reducing manual editing and human errors.