Swiss Federal Research Institute WSL, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland.
Mol Ecol Resour. 2013 Jan;13(1):128-34. doi: 10.1111/1755-0998.12033. Epub 2012 Nov 26.
The use of procedures for the automated scoring of amplified fragment length polymorphisms (AFLP) fragments has recently increased. Corresponding software does not only automatically score the presence or absence of AFLP fragments, but also allows an evaluation of how different settings of scoring parameters influence subsequent population genetic analyses. In this study, we used the automated scoring package rawgeno to evaluate how five scoring parameters influence the number of polymorphic bins and estimates of pairwise genetic differentiation between populations (F(st)). Steps were implemented in r to automatically run the scoring process in rawgeno for a set of different parameter combinations. While we found the scoring parameters minimum bin width and minimum number of samples per bin to have only weak influence on pairwise F(st) values, maximum bin width and bin reproducibility had much stronger effects. The minimum average bin fluorescence scoring parameter affected F(st) values in an only moderate way. At a range of scoring parameters around the default settings of rawgeno, the number of polymorphic bins as well as pairwise F(st) values stayed rather constant. This study thus shows the particularities of AFLP scoring, be it either manual or automatical, can have profound effects on subsequent population genetic analysis.
最近，使用程序自动评分扩增片段长度多态性（AFLP）片段的方法有所增加。相应的软件不仅可以自动评分 AFLP 片段的存在或缺失，还可以评估评分参数设置的不同如何影响后续的种群遗传分析。在这项研究中，我们使用了自动评分软件 rawgeno 来评估五个评分参数如何影响多态性分箱的数量和种群间成对遗传分化（Fst）的估计值。在 r 中实现了步骤，以便为一组不同的参数组合自动运行 rawgeno 中的评分过程。虽然我们发现评分参数最小分箱宽度和每个分箱最小样本数对成对 Fst 值的影响较弱，但最大分箱宽度和分箱重现性的影响要强得多。最小平均分箱荧光评分参数对 Fst 值的影响中等。在 rawgeno 默认设置周围的一系列评分参数范围内，多态性分箱的数量以及成对 Fst 值保持相当稳定。因此，本研究表明 AFLP 评分的特殊性，无论是手动还是自动，都可能对后续的种群遗传分析产生深远的影响。
