Normalization and binning of historical and multi-source microsatellite data: overcoming the problems of allele size shift with allelogram

Abstract

Microsatellite allele data have long been plagued by size shifts that can at best make it difficult to accurately assign genotypes to allele products, and at worse can cause whole batches of data from different instruments, dates or laboratories to be incorrectly assigned. Although modern genotyping technology (capillary electrophoresis) has overcome many of these problems, concern remains regarding the consistency of scores within a laboratory over time and between laboratories when combining data from multiple sources into a single analysis. There remain a large number of laboratories using older technologies or combining data from multiple sources. In addition, thousands of data sets that could potentially be expanded as samples become available are generally regarded as unusable because of the effort that would be required to validate congruence of genotypes from old and new data sets. We present methods to normalize and bin alleles from multiple data sources using a relatively small set of controls and the freely available program allelogram.