XML for input into BEAST with increasing temporal sampling range

XML were created in BEAUti v1.5.4 for analysis in BEAST v1.7.4, as described in Hedge et al. 2013 (Drummond & Rambaut, 2007, Drummond et al. 2012). Each file specifies the EpiFlu accession numbers of the segments concatenated into the whole-genome sequences and used as input data for the analysis (http://platform.gisaid.org/). Each file is labelled with the name of the last month during which sequences included in the analysis were sampled and the growth modelled employed. For example, June.exp.xml comprises EpiFlu accession numbers of sequences sampled from the start of the pandemic until June 2009 and specifies the use of an exponential growth model. Details of the substitution, clock and coalescent models are provided along with the priors placed on the model parameters. An additional block implementing the estimation of the marginal likelihood of the model using path-sampling is given after the Markov chain Monte Carlo (MCMC) block (Baele et al. 2012).本数据集所用XML文件均基于BEAUti v1.5.4生成，用于BEAST v1.7.4的分析流程，具体方法参见Hedge等人2013年的研究（以及Drummond与Rambaut 2007年、Drummond等人2012年的文献）。每个XML文件均列明了拼接为全基因组序列的各片段的EpiFlu登录号，并将其作为本次分析的输入数据（数据来源：http://platform.gisaid.org/）。所有文件均以本次分析纳入的序列的最后采样月份以及所采用的生长模型进行命名。例如，June.exp.xml包含了自疫情暴发伊始至2009年6月采样的序列的EpiFlu登录号，并指定采用指数生长模型。文件中还提供了替换模型、分子钟模型以及溯祖模型的详细参数设置，以及各模型参数的先验分布设定。在马尔可夫链蒙特卡洛（Markov Chain Monte Carlo, MCMC）模块之后，还增设了一个用于通过路径采样法估算模型边际似然的模块（Baele等人2012年）。