绘制组间分类数据的共同元素

Question

我有一份数据，看起来像这样（实际数据有4000行）：

tt <- structure(list(X = c("k__Fungi;p__Ascomycota;c__Pezizomycotina_cls_Incertae_sedis;o__Pezizomycotina_ord_Incertae_sedis;f__Pezizomycotina_fam_Incertae_sedis;g__Vermispora", 
"k__Fungi;p__Ascomycota;c__Saccharomycetes;o__Saccharomycetales;f__Saccharomycetales_fam_Incertae_sedis;g__Candida", 
"k__Fungi;p__Ascomycota;c__Sordariomycetes;o__Branch06;f__unidentified;g__unidentified", 
"k__Bacteria;p__Proteobacteria;c__Betaproteobacteria;o__Burkholderiales;f__Oxalobacteraceae;g__Massilia", 
"k__Bacteria;p__Actinobacteria;c__Actinobacteria;o__Actinomycetales;f__Pseudonocardiaceae;g__Amycolatopsis", 
"k__Bacteria;p__Armatimonadetes;c__Armatimonadia;o__FW68;f__;g__", 
"k__Bacteria;p__Proteobacteria;c__Deltaproteobacteria;o__Syntrophobacterales;f__Syntrophobacteraceae;g__", 
"k__Fungi;p__Ascomycota;c__Dothideomycetes;o__Pleosporales;f__Leptosphaeriaceae;g__Leptosphaeria", 
"D_0__Bacteria;D_1__WPS-2;D_2__metagenome;D_3__metagenome;D_4__metagenome;D_5__metagenome", 
"D_0__Bacteria;D_1__Patescibacteria;D_2__Saccharimonadia;D_3__Saccharimonadales;D_4__Saccharimonadaceae;D_5__Candidatus Saccharibacteria bacterium RAAC3_TM7_1", 
"D_0__Bacteria;D_1__Chloroflexi;D_2__AD3;D_3__uncultured bacterium;D_4__uncultured bacterium;D_5__uncultured bacterium", 
"k__Fungi;p__Chytridiomycota;c__Chytridiomycetes;o__Rhizophydiales;f__Rhizophydiales_fam_Incertae_sedis;g__Coralloidiomyces", 
"k__Bacteria;p__Bacteroidetes;c__Cytophagia;o__Cytophagales;f__Cytophagaceae;g__Pontibacter", 
"D_0__Bacteria;D_1__WS2;D_2__uncultured soil bacterium;D_3__uncultured soil bacterium;D_4__uncultured soil bacterium;D_5__uncultured soil bacterium", 
"k__Fungi;p__Basidiomycota;c__Agaricomycetes;o__Thelephorales;f__Thelephoraceae;g__Amaurodon", 
"D_0__Bacteria;D_1__Patescibacteria;D_2__Parcubacteria;D_3__Candidatus Kaiserbacteria;D_4__metagenome;D_5__metagenome", 
"k__Bacteria;p__Proteobacteria;c__Betaproteobacteria;o__Burkholderiales;f__Oxalobacteraceae;g__Janthinobacterium", 
"k__Bacteria;p__Bacteroidetes;c__Bacteroidia;o__Bacteroidales;__;__", 
"D_0__Bacteria;D_1__Bacteroidetes;D_2__Bacteroidia;D_3__Sphingobacteriales;D_4__Sphingobacteriaceae;D_5__Pedobacter", 
"k__Bacteria;p__Acidobacteria;c__Acidobacteriia;o__Acidobacteriales;f__Koribacteraceae;g__", 
"D_0__Bacteria;D_1__Planctomycetes;D_2__Planctomycetacia;D_3__Pirellulales;D_4__Pirellulaceae;D_5__Pirellula", 
"k__Fungi;p__Ascomycota;c__Pezizomycetes;o__Pezizales;f__Sarcosomataceae;g__Urnula", 
"D_0__Bacteria;D_1__Proteobacteria;D_2__Alphaproteobacteria;D_3__Azospirillales;D_4__uncultured;D_5__uncultured bacterium", 
"k__Fungi;p__Basidiomycota;c__Agaricomycetes;o__Polyporales;f__Meruliaceae;g__Uncobasidium", 
"D_0__Bacteria;D_1__Dependentiae;D_2__Babeliae;D_3__Babeliales;D_4__Vermiphilaceae;__", 
"D_0__Bacteria;D_1__Proteobacteria;D_2__Alphaproteobacteria;D_3__Caulobacterales;D_4__Caulobacteraceae;D_5__Caulobacter", 
"k__Fungi;p__Mortierellomycota;__;__;__;__", "k__Fungi;p__Ascomycota;c__Dothideomycetes;o__Pleosporales;f__Dictyosporiaceae;g__Dictyosporium", 
"k__Bacteria;p__Acidobacteria;c__;o__;f__;g__", "D_0__Bacteria;D_1__Patescibacteria;D_2__Gracilibacteria;D_3__Absconditabacteriales (SR1);__;__", 
"k__Fungi;p__Basidiomycota;c__Agaricomycetes;o__Agaricales;f__Entolomataceae;g__Clitopilus", 
"k__Bacteria;p__Proteobacteria;c__Betaproteobacteria;o__Burkholderiales;f__Burkholderiaceae;g__", 
"D_0__Bacteria;D_1__Armatimonadetes;D_2__Fimbriimonadia;D_3__Fimbriimonadales;D_4__Fimbriimonadaceae;D_5__Armatimonadetes bacterium 55-13", 
"k__Fungi;p__Basidiomycota;c__Agaricomycetes;o__Agaricales;f__Bolbitiaceae;g__Agrocybe", 
"k__Fungi;p__Ascomycota;c__Leotiomycetes;o__Helotiales;f__Helotiales_fam_Incertae_sedis;__", 
"D_0__Bacteria;D_1__Cyanobacteria;D_2__Oxyphotobacteria;D_3__Oxyphotobacteria Incertae Sedis;D_4__Unknown Family;D_5__Leptolyngbya ANT.L52.2", 
"k__Bacteria;p__Chloroflexi;c__Chloroflexi;o__[Roseiflexales];f__[Kouleothrixaceae];__", 
"D_0__Bacteria;D_1__Proteobacteria;D_2__Alphaproteobacteria;D_3__Rhizobiales;D_4__Hyphomicrobiaceae;__", 
"k__Fungi;p__Ascomycota;c__Eurotiomycetes;o__Onygenales;f__Onygenales_fam_Incertae_sedis;g__Spiromastix", 
"k__Fungi;p__Basidiomycota;c__Tremellomycetes;o__Tremellales;f__Bulleribasidiaceae;g__Vishniacozyma", 
"k__Fungi;p__Basidiomycota;c__Tremellomycetes;o__Tremellales;f__Bulleribasidiaceae;g__Bulleribasidium", 
"k__Fungi;p__Ascomycota;c__Dothideomycetes;o__Pleosporales;f__Corynesporascaceae;g__Corynespora", 
"k__Fungi;p__Ascomycota;c__Eurotiomycetes;o__Chaetothyriales;f__Cyphellophoraceae;g__Cyphellophora", 
"k__Bacteria;p__Proteobacteria;c__Gammaproteobacteria;o__Xanthomonadales;f__Sinobacteraceae;__", 
"D_0__Bacteria;D_1__Proteobacteria;D_2__Deltaproteobacteria;D_3__SAR324 clade(Marine group B);__;__", 
"k__Fungi;p__Ascomycota;c__Leotiomycetes;o__Leotiomycetes_ord_Incertae_sedis;f__Myxotrichaceae;g__Oidiodendron", 
"k__Bacteria;p__Proteobacteria;c__Deltaproteobacteria;o__Myxococcales;f__Cystobacteraceae;g__Cystobacter", 
"D_0__Bacteria;D_1__Patescibacteria;D_2__Parcubacteria;D_3__Candidatus Kaiserbacteria;D_4__uncultured bacterium;D_5__uncultured bacterium", 
"k__Fungi;p__Chytridiomycota;c__Spizellomycetes;o__Spizellomycetales;f__Spizellomycetaceae;g__Spizellomyces", 
"k__Bacteria;p__Armatimonadetes;c__Chthonomonadetes;o__SJA-22;f__;g__"
), Reject.null.hypothesis = c("False", "False", "False", "False", 
"False", "False", "False", "False", "False", "False", "False", 
"False", "False", "False", "False", "False", "False", "False", 
"False", "False", "False", "False", "False", "False", "False", 
"False", "False", "False", "False", "False", "False", "False", 
"False", "False", "False", "False", "False", "False", "False", 
"False", "False", "False", "False", "False", "False", "False", 
"False", "False", "False", "False"), W = c(3L, 57L, 89L, 0L, 
3L, 0L, 6L, 33L, 27L, 28L, 1L, 3L, 5L, 1L, 3L, 1L, 0L, 10L, 1L, 
15L, 26L, 1L, 1L, 4L, 1L, 0L, 6L, 15L, 0L, 23L, 6L, 1L, 45L, 
4L, 0L, 1L, 5L, 1L, 45L, 3L, 7L, 192L, 53L, 0L, 2L, 8L, 0L, 18L, 
7L, 0L), Type = structure(c(4L, 6L, 6L, 2L, 1L, 1L, 2L, 9L, 5L, 
5L, 7L, 4L, 1L, 7L, 3L, 7L, 1L, 1L, 8L, 1L, 5L, 4L, 7L, 9L, 8L, 
7L, 10L, 10L, 1L, 5L, 10L, 2L, 5L, 9L, 4L, 7L, 1L, 8L, 6L, 9L, 
9L, 4L, 6L, 1L, 7L, 3L, 1L, 5L, 9L, 2L), .Label = c("Soil_16S_Monmouth_ancom.csv", 
"Soil_16S_Urbana_ancom.csv", "Soil_ITS_Monmouth_ancom.csv", "Soil_ITS_urbana_ancom.csv", 
"Rhizoshpere_16S_U_and_M_together_ancom.tsv", "Rhizoshpere_ITS_U_and_M_together_ancom.tsv", 
"Rhizoshpere_16S_Monmouth_only_ancom.tsv", "Rhizoshpere_16S_Urbana_only_ancom.tsv", 
"Rhizoshpere_ITS_Monmouth_only_ancom.tsv", "Rhizoshpere_ITS_Urbana_only_ancom.tsv"
), class = "factor"), microbiome_type = c("Soil", "Rhizosphere", 
"Rhizosphere", "Soil", "Soil", "Soil", "Soil", "Rhizosphere", 
"Rhizosphere", "Rhizosphere", "Rhizosphere", "Soil", "Soil", 
"Rhizosphere", "Soil", "Rhizosphere", "Soil", "Soil", "Rhizosphere", 
"Soil", "Rhizosphere", "Soil", "Rhizosphere", "Rhizosphere", 
"Rhizosphere", "Rhizosphere", "Rhizosphere", "Rhizosphere", "Soil", 
"Rhizosphere", "Rhizosphere", "Soil", "Rhizosphere", "Rhizosphere", 
"Soil", "Rhizosphere", "Soil", "Rhizosphere", "Rhizosphere", 
"Rhizosphere", "Rhizosphere", "Soil", "Rhizosphere", "Soil", 
"Rhizosphere", "Soil", "Soil", "Rhizosphere", "Rhizosphere", 
"Soil")), row.names = c(3771L, 3792L, 3806L, 2839L, 1913L, 2004L, 
2956L, 3311L, 1619L, 1008L, 535L, 4785L, 2087L, 1625L, 4564L, 
977L, 2836L, 2059L, 510L, 1706L, 1090L, 3750L, 1143L, 4514L, 
848L, 1158L, 4873L, 3268L, 1658L, 952L, 4299L, 2792L, 408L, 4279L, 
3610L, 827L, 2217L, 1275L, 3573L, 4713L, 4704L, 3258L, 3499L, 
3018L, 1487L, 3668L, 2913L, 982L, 4806L, 2013L), class = "data.frame")

我想绘制X与微生物群落类型的图表，以显示在土壤和根际中共同存在的X。也就是说，我想查看哪些细菌和真菌群落（来自列X）在Rhizosphere和Soil微生物群落类型之间被共享（以及未被共享）。

我尝试了以下类似的内容，但它没有给我想要的结果：

ggplot(tt, aes(microbiome_type, X)) +
  geom_jitter(aes(color = Type), size = 0.9)+
  ggpubr::color_palette("jco")+
  ggpubr::theme_pubclean() +
  theme(axis.text.y=element_blank()) +
  facet_wrap(~Reject.null.hypothesis) 

请问有没有更好的方法来绘制这种类型的数据？

- MAPK

Answer 1

我不从事微生物学，所以我不知道为了正确回答你的问题应该进行何种精确级别的分析。然而，目前你的数据结构似乎无法很好地回答你的问题。例如，我预计一些真菌和细菌物种在X中会出现多次，但是X是描述特定微生物群落的因素，在其中会丢失成员级别信息，因为它被编码在因素水平中。

我的建议是将X拆分为各个分类单元，对于这些分类单元，可视化它们是否存在于土壤或根际中。

new_x <- strsplit(tt$X, ";")

# Recombine with original information, you might get a warning about rownames
newdat <- lapply(seq_along(new_x), function(i) {
  cbind(X = new_x[[i]], tt[i,-1])
})
newdat <- do.call(rbind, newdat)

ggplot(newdat, aes(microbiome_type, X, group = X)) +
  geom_point(aes(colour = Type)) +
  geom_line() +
  facet_wrap(~ Reject.null.hypothesis)

现在很容易发现土壤和根际中的物种是什么。

- teunbrand