Department of Animal and Plant Sciences, University of Sheffield, Sheffield S10 2TN, UK.
Mol Ecol Resour. 2008 Jul;8(4):725-35. doi: 10.1111/j.1755-0998.2007.02073.x.
Amplified fragment length polymorphism (AFLP) fingerprint data are now commonly collected using DNA sequencers. AFLP genotypes are still often scored by eye from such data - a time-consuming, error-prone and subjective process. We present a semi-automated method of genotyping sequencer-collected AFLPs at predefined fragment locations (loci) within the fingerprint. Our method uses thresholds of AFLP-polymerase chain reaction-product fluorescence intensity (peak height) in order to: (i) exclude AFLP loci that are likely to contribute high rates of error to data sets, and (ii) determine the AFLP phenotype (fragment presence or absence) at the retained loci. Error rate analysis is an integral part of this process and is used to determine optimal thresholds that minimize genotyping error, while maximizing the numbers of retained loci. We show that application of this method to a large AFLP data set allows genotype calls that are rapid, objective and repeatable, facilitating the extraction of reliable genotype data for molecular ecological studies.
扩增片段长度多态性（AFLP）指纹图谱数据现在通常使用 DNA 测序仪进行收集。AFLP 基因型仍然经常通过肉眼从这些数据中进行评分——这是一个耗时、容易出错且主观的过程。我们提出了一种在指纹图谱中预设片段位置（基因座）上对测序仪收集的 AFLP 进行半自动化基因分型的方法。我们的方法使用 AFLP-聚合酶链反应产物荧光强度（峰高）的阈值，以：（i）排除可能导致数据集错误率高的 AFLP 基因座，以及（ii）确定保留基因座处的 AFLP 表型（片段存在或不存在）。误差率分析是该过程的一个组成部分，用于确定最佳阈值，以最大限度地减少基因分型误差，同时最大限度地保留基因座数量。我们表明，将该方法应用于大型 AFLP 数据集可以快速、客观和可重复地进行基因型调用，为分子生态学研究提取可靠的基因型数据提供了便利。
