-
Notifications
You must be signed in to change notification settings - Fork 0
/
ASLeafITS.rmd
194 lines (140 loc) · 6.12 KB
/
ASLeafITS.rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
---
title: "ASLeaf ITS"
author: "JReceveur"
output:
html_document:
code_folding: hide
toc: true
toc_depth: 3
toc_float: true
editor_options:
chunk_output_type: console
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(fig.width=14, fig.height=10)
knitr::opts_chunk$set(fig.align="center")
```
# Import{.tabset}
## Sample Overview
```{r,warning=F,include=F}
#From https://gist.github.com/grabear/018e86413b19b62a6bb8e72a9adba349
#Parse Silva function
parse_taxonomy_silva_138 <- function(char.vec){
# Use default to assign names to elements in case problem with greengenes prefix
char.vec = parse_taxonomy_default(char.vec)
# Check for unassigned taxa
if (char.vec["Rank1"] == "Unassigned") {
char.vec <- c(Rank1="d__Unassigned", Rank2="d__Unassigned", Rank3="d__Unassigned", Rank4="d__Unassigned",
Rank5="d__Unassigned", Rank6="d__Unassigned", Rank7="d__Unassigned")
}
# Define the meaning of each prefix according to SILVA taxonomy
Tranks = c(Rank1="Kingdom", Rank2="Phylum", Rank3="Class", Rank4="Order", Rank5="Family", Rank6="Genus", Rank7="Species")
# Check for prefix using regexp, warn if there were none. trim indices, ti
ti = grep("[[:alpha:]]\\_\\_", char.vec)
if( length(ti) == 0L ){
warning(
"No silva prefixes were found. \n",
"Consider using parse_taxonomy_delfault() instead if true for all OTUs. \n",
"Dummy ranks may be included among taxonomic ranks now."
)
# Will want to return without further modifying char.vec
taxvec = char.vec
# Replace names of taxvec according to prefix, if any present...
} else {
# Format character vectors for Ambiguous taxa
if( length(ti) < 7 ){
for (key in names(char.vec)) {
if ( char.vec[key] == "Ambiguous_taxa" ) {
tax_no <- (as.numeric(substr(key, 5, 5)) - 1)
char.vec[key] = sprintf("d__Ambiguous_taxa", tax_no)
}
}
# Reset the trimmed indicies if Ambiguous taxa
ti = grep("[[:alpha:]]\\_\\_", char.vec)
}
# Remove prefix using sub-"" regexp, call result taxvec
taxvec = gsub("[[:alpha:]]\\_\\_", "", char.vec)
# Define the ranks that will be replaced
repranks = Tranks[substr(char.vec[ti], 1, 3)]
# Replace, being sure to avoid prefixes notK present in Tranks
names(taxvec)[ti[!is.na(repranks)]] = repranks[!is.na(repranks)]
}
return(taxvec)
}
library(vegan)
library(ggplot2)
library(plyr)
library(dplyr)
library(phyloseq)
library(randomForest)
library(knitr)
set.seed(10)
cbPalette <- c("#999999", "#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#000000","#CC79A7") #Build a user defined colorblind palette
theme_set(theme_bw(base_size = 18)+theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())) #Set font size and change default plot theme
biom=import_biom("ASLeafJan2021ITS_6k.biom",parseFunction=parse_taxonomy_silva_138) #Import biom file which contains taxa names and abundances
colnames(tax_table(biom))<-c(Rank1="Kingdom", Rank2="Phylum", Rank3="Class", Rank4="Order", Rank5="Family", Rank6="Genus", Rank7="Species")
metadata=read.table("LauraMetadata9Dec2020.txt",header = TRUE)
metadata$Date = factor(metadata$Date, levels = c("3_1","24_1","14_2","7_3","29_3","19_4")) #fixes x-axis labels
metadata$Paired<-paste0(metadata$Site,metadata$Date)
tree=read_tree("ASLeafTreeDec2020.nwk")
sampdat=sample_data(metadata)
sample_names(sampdat)=metadata$ID # If using fungal data change ID to FungalID
physeq=merge_phyloseq(biom,sampdat,tree)
physeq<-filter_taxa(physeq, function (x) {sum(x > 0) > 1}, prune=TRUE)#Remove zeros
physeq<-subset_samples(physeq, Site!="<NA>") # Remove Control
physeq
#physeq
GPr = transform_sample_counts(physeq, function(x) x / sum(x) ) #transform samples based on relative abundance
PhylumAll=tax_glom(GPr, "Phylum")
PhylumLevel = filter_taxa(PhylumAll, function(x) mean(x) > 3e-2, TRUE) #filter out any taxa lower than 3%
FamilyAll=tax_glom(GPr,"Family")
FamilyLevel = filter_taxa(FamilyAll, function(x) mean(x) > 3e-2, TRUE) #filter out any taxa lower than 3%
GenusAll=tax_glom(GPr,"Genus")
GenusLevel = filter_taxa(GenusAll, function(x) mean(x) > 2e-2, TRUE) #filter out any taxa lower than 2%
```
Unique fungal ASVs between leaf types
```{r}
df <- psmelt(physeq)
head(df)
length(unique(df$OTU))
OakSubset<-subset(df,Treatment=="Q")
OakSubset<-subset(OakSubset,Abundance!=0)
length(unique(OakSubset$OTU))
ChestnutSubset<-subset(df,Treatment=="C")
ChestnutSubset<-subset(ChestnutSubset,Abundance!=0)
length(unique(ChestnutSubset$OTU))
length(intersect(OakSubset$OTU,ChestnutSubset$OTU)) #2228
length(setdiff(ChestnutSubset$OTU,OakSubset$OTU)) #311 ASVs in Chestnut but not in Oak
length(setdiff(OakSubset$OTU,ChestnutSubset$OTU)) # 454 ASVs in Oak but not in Chestnut
## Unique ASVs by day
#########
# Day 21
#########
Timepoint1<-subset_samples(physeq,Days==21)
df1 <- psmelt(Timepoint1)
OakSubset1<-subset(df1,Treatment=="Q")
OakSubset1<-subset(OakSubset1,Abundance!=0)
length(unique(OakSubset1$OTU)) #1106
ChestnutSubset1<-subset(df1,Treatment=="C")
ChestnutSubset1<-subset(ChestnutSubset1,Abundance!=0)
length(unique(ChestnutSubset1$OTU)) #1108
length(intersect(OakSubset1$OTU,ChestnutSubset1$OTU)) #594
length(union(OakSubset1$OTU,ChestnutSubset1$OTU)) #1520
length(setdiff(ChestnutSubset1$OTU,OakSubset1$OTU)) #414 ASVs in Chestnut but not in Oak
length(setdiff(OakSubset1$OTU,ChestnutSubset1$OTU)) # 512 ASVs in Oak but not in Chestnut
#########
# Day 126
#########
Timepoint126<-subset_samples(physeq,Days==126)
df126 <- psmelt(Timepoint126)
OakSubset126<-subset(df126,Treatment=="Q")
OakSubset126<-subset(OakSubset126,Abundance!=0)
length(unique(OakSubset126$OTU)) #1530
ChestnutSubset126<-subset(df126,Treatment=="C")
ChestnutSubset126<-subset(ChestnutSubset126,Abundance!=0)
length(unique(ChestnutSubset126$OTU)) #1534
length(intersect(OakSubset126$OTU,ChestnutSubset126$OTU)) #1095
length(union(OakSubset126$OTU,ChestnutSubset126$OTU)) #1969
length(setdiff(ChestnutSubset126$OTU,OakSubset126$OTU)) #439 ASVs in Chestnut but not in Oak at day 126
length(setdiff(OakSubset126$OTU,ChestnutSubset126$OTU)) # 435 ASVs in Oak but not in Chestnut at day 126
```