Supplementary Information for “ARK: Aggregation of Reads by K-means for Estimation of Bacterial Community Composition”.

This supporting information is available online. This supplementary material is included to address eight major points: To compare ARK with the best performing bacterial community composition method to date, called BEBaC [8]. BEBaC employs a Bayesian estimation clustering framework along-with a stochastic search and sequence alignment. To investigate the important question of finding the number of regions Q in ARK. To independently verify ARK in two different geographic regions ((1) Sweden and Finland, and (2) USA) and also using different datasets. To detail genera-level reconstructions of ARK SEK, ARK Quikr, and RDP’s NBC. To detail the primers used to obtain the data in the main text. To demonstrate the results are qualitatively independent of the error correction method chosen. To detail the effect of changing the k-mer size. To investigate the behavior of each method when sister taxa are excluded from the training database. To compare ARK with the best performing bacterial community composition method to date, called BEBaC [8]. BEBaC employs a Bayesian estimation clustering framework along-with a stochastic search and sequence alignment. To investigate the important question of finding the number of regions Q in ARK. To independently verify ARK in two different geographic regions ((1) Sweden and Finland, and (2) USA) and also using different datasets. To detail genera-level reconstructions of ARK SEK, ARK Quikr, and RDP’s NBC. To detail the primers used to obtain the data in the main text. To demonstrate the results are qualitatively independent of the error correction method chosen. To detail the effect of changing the k-mer size. To investigate the behavior of each method when sister taxa are excluded from the training database. (PDF)

本支持信息可在线获取。本补充材料旨在涵盖八大核心内容：
1. 将ARK与当前性能最优的细菌群落组成分析方法BEBaC[8]进行对比，BEBaC采用贝叶斯估计聚类框架，结合随机搜索与序列比对算法。
2. 探究ARK中区域数量Q的确定这一关键科学问题。
3. 在(1)瑞典与芬兰、(2)美国两个不同地理区域，并采用不同数据集的前提下，独立验证ARK的性能表现。
4. 详细阐述ARK SEK、ARK Quikr以及RDP的NBC（朴素贝叶斯分类器）的属级重建结果。
5. 详述正文中实验数据获取所用的引物序列。
6. 证明研究结果在定性层面与所选用的序列纠错方法无关。
7. 详细分析k-mer（k元组）长度变化所带来的影响。
8. 探究当训练数据库剔除姊妹类群时，各方法的运行表现。
本补充材料为PDF格式。