Microsatellite allele sizing: difference between automated capillary electrophoresis and manual technique.

By comparing data collected with different automated sequencers and a manual technique (fragment separation in a silver-stained polyacrylamide gel), we found strong discrepancies in allele size of microsatellite loci. To quantify the sizing bias generated by automated capillary electrophoresis, we typed 51 alleles at seven loci and found that differences between actual (manual) and called (automated) sizing were inversely related to locus size. This result seems independent of the fluorescent dye but might be due to different migration patterns of the size standard and the microsatellite loci. Thus, it is essential to distinguish between actual (that can only he confirmed by sequencing) and called (obtained with automated sequencer) allele sizes. To enable the comparison of data collected by different laboratories on different instruments, the greatest attention should be paid to material and protocol descriptions used for allele sizing, and reference standard DNA genotypes should he shared between collaborating laboratories. Without these precautions, scoring errors in allele size might result in important misleading conclusions.