R语言：microeco:一个用于微生物群落生态学数据挖掘的R包,第四：trans_beta class

trans_beta class：利用trans_beta类可以变换和绘制beta分集的距离矩阵。该类中涉及到beta多样性的分析主要包括排序、群距、聚类和方差分析。我们首先使用PCoA显示排序。

> dataset$cal_betadiv()
The result is stored in object$beta_diversity ...
> t1 <- trans_beta$new(dataset = dataset, group = "Group", measure = "bray")
> t1$cal_ordination(ordination = "PCoA")
The ordination result is stored in object$res_ordination ...
> tmp <- t1$res_ordination$scores
> t2 <- trans_env$new(dataset = dataset, add_data = tmp[, 1:2])
Env data is stored in object$data_env ...
> t2$cal_diff(group = "Group", method = "anova")
The result is stored in object$res_diff ...
Warning message:
程辑包‘agricolae’是用R版本4.3.3 来建造的  #这里需要安装agricolae包，直接install就行。
> t2
trans_env object:
Env table have 2 variables: PCo1,PCo2

> p1 <- t1$plot_ordination(plot_color = "Group", plot_shape = "Group", plot_type = c("point", "ellipse"))

#然后我们绘制并比较群距。

> t1$cal_group_distance()

> t1$plot_group_distance(distance_pair_stat = TRUE)

#这里应该会有差异比较的，但是却没有。示例如下，不知道有什么问题，等到有需求我会解决这个问题。

#计算和绘制组之间的样本距离

> t1$cal_group_distance(within_group = FALSE)

> t1$plot_group_distance(distance_pair_stat = TRUE)

聚类图也是一种常用的方法。

> t1$plot_clustering(group = "Group", replace_name = c("Saline", "Type"))

#perMANOVA常用于组间距离的差异检验。

> t1$cal_manova(cal_manova_all = TRUE)
The result is stored in object$res_manova ...
> t1$res_manova
Permutation test for adonis under reduced model
Terms added sequentially (first to last)
Permutation: free
Number of permutations: 999

adonis2(formula = use_formula, data = metadata, cal_manova_all = TRUE)
         Df SumOfSqs      R2      F Pr(>F)    
Group     2   6.1207 0.19553 10.573  0.001 ***
Residual 87  25.1822 0.80447                  
Total    89  31.3029 1.00000                  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

manova表示每一对组

> t1$cal_manova(cal_manova_paired = TRUE)
The result is stored in object$res_manova ...
> t1$res_manova
Permutation test for adonis under reduced model
Terms added sequentially (first to last)
Permutation: free
Number of permutations: 999

adonis2(formula = use_formula, data = metadata, cal_manova_paired = TRUE)
         Df SumOfSqs      R2      F Pr(>F)    
Group     2   6.1207 0.19553 10.573  0.001 ***
Residual 87  25.1822 0.80447                  
Total    89  31.3029 1.00000                  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

# manova用于指定组集:这里是“group + Type”
> t1$cal_manova(cal_manova_set = "Group + Type")
The result is stored in object$res_manova ...
> t1$res_manova
Permutation test for adonis under reduced model
Terms added sequentially (first to last)
Permutation: free
Number of permutations: 999

adonis2(formula = use_formula, data = metadata, cal_manova_set = "Group + Type")
         Df SumOfSqs      R2      F Pr(>F)    
Group     2   6.1207 0.19553 10.573  0.001 ***
Residual 87  25.1822 0.80447                  
Total    89  31.3029 1.00000                  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

这个包的改动有些大，以前的代码不能直接跑了。我发觉，学习这个包还是要基于需求啊，否则会很快忘掉。以前我使用vegan包分析+ggplot2包做PCOA分析，等把这个包分享完了，我会再分享其它的R语言。