Metastats Analysis ( !!rather slow, not recommended )(R语言版本metastat分析)
显著性差异分析[1](Differentially Abundant Features)根据得到的群落丰度数据,运用严格的统计学方法可以检测两组微生物群落中表现出丰度差异的分类,进行稀有频率数据的多重假设检验和假发现率(FDR)分析可以评估观察到的差异的显著性。分析可选择输入门、纲、目、科及属等不同分类学水平。
输入:
OTU Table文件:
OTU ID Bio1 Bio2 Bio3 Bio4 Bio5 Bio6 Bio7 Bio8 Bio9 Bio10
OTU1 0 0 0 0 0 6 34 104 367 254
OTU2 52 335 18 49 0 0 0 0 0 0
OTU3 0 0 0 0 5 0 0 0 0 0
样品分组信息表:
!!注:保证每个分组包含的样品数量不少于2(>=2),即每个分组都需要有重复样。
Bio1 group1
Bio2 group1
Bio3 group1
Bio4 group1
Bio5 group2
Bio6 group2
Bio7 group2
Bio8 group2
其他参数默认。
输出:
组间丰度差异分析结果:
group2(Bio5,Bio6,Bio7,Bio8) vs. group1(Bio1,Bio2,Bio3,Bio4)
Name Mean(group1) Variance(group1) std.err(group1) Mean(group2) Variance(group2) std.err(group2) pvalue qvalue
OTU1 0.00112258549755647 1.77588913967284e-06 0.000666312452921458 0 0 0 0.0435342163355408 0.095923873844819
OTU2 0 0 0 0.00341202642922868 2.22085684102574e-05 0.0023562983899677 0.0992406181015453 0.120738713533526
OTU3 4.16944629753169e-05 6.95371297120028e-09 4.16944629753169e-05 0 0 0 0.309490066225166 0.136096773290994
OTU4 0 0 0 0.00029056992067046 4.12213524125127e-08 0.000101515211190876 0.0018841059602649 0.0264842802252221
OTU5 0 0 0 1.25760853161628e-05 6.32631687517621e-10 1.25760853161628e-05 0.503020774582078 0.215392125605884
注:mean:均值; variance:方差; standard:标准差;
p value (an individual measure of the false positive rate) 假阳性概率值,是统计学中常用的判定值,一般来说P value<0.05 时差异显著;
q value (an individual measurement of the false discovery rate) 假发现率评估值,指本次计算可信度。
分析模块引用了matastats[2](http://metastats.cbcb.umd.edu/)。
相关文献如下所示:
[1]Tingting Wang, Guoxiang Cai, et al. Structural segregation of gut microbiota between colorectal cancer patients and healthy volunteers. The ISME Journal advance online publication, 18 August 2011; doi:10.1038/ismej.2011.109.
[2]White, J.R., Nagarajan, N. & Pop, M. Statistical methods for detecting differentially abundant features in clinical metagenomic samples. PLoS Comput Biol 5, e1000352 (2009).