Microsatellites are at the moment the most used marker to map quantitative trait and disease loci (QTLs). An alternative to genotype single individuals is amplifying just in one reaction a group of animals which ones make up a pool of DNA. We have quantified the precision of this technique to estimate allele frequencies for four microsatellites in pools from three to fifty individuals. The PCR product was analysed through capillary electrophoresis using an ABI‑310 sequencer and the electropherograms were corrected for stutter bands and adjusted for preferential amplification. Both are artifactual PCR products, which are present in the majority of the microsatellites.
Peak height and area were assessed with the objective of determining allele frequencies. V_T is the technical error variance used to estimate the accuracy of prediction of allele frequency and it was lower for peak height than peak area in every used marker. The average V_T was 0.001, that is, any allele frequency x is in the range x ± 6,3% for 95% confidence interval. The correlation between estimated all de frequencies from pools and allele frequencies based on individual genotyping was always higher than 0.93. No V_T increment was observed with the increasing number of individuals included in the pool.