-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathLat_Clean_Data.Rmd
228 lines (194 loc) · 11.3 KB
/
Lat_Clean_Data.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
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
---
title: "This code is for the analysis of 16S Sequencing Data from a latitudinal study along the Eastern Coast of Australia"
author: "Nathan Williams"
date: "22/10/2020"
output: html_document
---
```{r Libraries}
#0.1 Set working directory
setwd("~/Dropbox/UTS/PhD/Projects/Chapter2_RoseBay/Data_Analysis/Data/16S/Data")
#0.2 Load dependencies
library("gridExtra")
library("vegan")
library("metacoder")
library("taxa")
library("phyloseq")
library("ggplot2")
library("dplyr")
library("readr")
library("stringr")
library("agricolae")
library("ape")
library("tidyverse")
library("tidyr")
library("ggpubr")
library("metagMisc")
library("microbiome")
```
```{r Asthetics}
#Theme for plot
NathanTheme <- list( #scale_x_discrete(limits=(locations)),
theme(axis.title.x = element_text(vjust = 0, hjust = 0.5, size = 16)),
theme(axis.title.y = element_text(vjust = 1, hjust = 0.5, size = 16)),
theme(axis.text=element_text(size=10)),
theme(axis.text.x = element_text(hjust=1, size = 14, angle = 90)),
theme(axis.text.y = element_text(hjust=1, size = 14, angle = 0)),
theme(panel.border = element_blank(), panel.grid.major = element_blank(),
panel.background = element_blank(),panel.grid.minor = element_blank(),
axis.line = element_line(colour = 'black')))
```
```{r Load and prepare files}
#0.1 Set working directory
setwd("~/Dropbox/UTS/PhD/Projects/Chapter1_Vibrio_latitude_study/Data/16S/Data")
#1.1 load in the data
asv_data <- read_csv("ASV_table.csv")
sample_data <- read_csv('SMD.csv')
#1.2 "select" your columns and then "spread" the data
asv_data <- asv_data %>% select(ASV,code2,abund_chim1)
asv_data <- asv_data %>% rename(SampleID = code2)
asv_data <- asv_data %>% rename(Abundance = abund_chim1)
asv_data <-asv_data %>% spread(key='SampleID', value='Abundance', fill=0)
names(asv_data) = gsub(pattern = "p1;", replacement = "", x = names(asv_data))
```
```{r Load in the taxa data}
setwd("~/Dropbox/UTS/PhD/Projects/Chapter1_Vibrio_latitude_study/Data/16S/Data")
tax.silva <- read_csv("Tax_silva.csv")
tax.species <- read_csv("Tax_species.csv")
Chloroplast_tax_data <- read_csv('PR2tax.csv')
tax_data <- tax.silva %>% left_join(tax.species, 'ASV')
```
```{r Set your threshold and remove your bootstrap of 20 and replace it with 49, this is the first step towards cleaning your data}
mythreshold <- 49
tax_data$tax.Kingdom <- ifelse(tax_data$boot.Kingdom <= mythreshold, NA, tax_data$tax.Kingdom)
tax_data$tax.Phylum <- ifelse(tax_data$boot.Phylum <= mythreshold, NA, tax_data$tax.Phylum)
tax_data$tax.Class <- ifelse(tax_data$boot.Class <= mythreshold, NA, tax_data$tax.Class)
tax_data$tax.Order <- ifelse(tax_data$boot.Order <= mythreshold, NA, tax_data$tax.Order)
tax_data$tax.Family <- ifelse(tax_data$boot.Family <= mythreshold, NA, tax_data$tax.Family)
tax_data$tax.Genus <- ifelse(tax_data$boot.Genus <= mythreshold, NA, tax_data$tax.Genus)
tax_data$tax.Kingdom <- forcats::fct_explicit_na(tax_data$tax.Kingdom, 'k_unassigned')
tax_data$tax.Phylum <- forcats::fct_explicit_na(tax_data$tax.Phylum, 'p_unassigned')
tax_data$tax.Class <- forcats::fct_explicit_na(tax_data$tax.Class, 'c_unassigned')
tax_data$tax.Order <- forcats::fct_explicit_na(tax_data$tax.Order, 'o_unassigned')
tax_data$tax.Family <- forcats::fct_explicit_na(tax_data$tax.Family, 'f_unassigned')
tax_data$Genus <- forcats::fct_explicit_na(tax_data$Genus, 'g_unassigned')
tax_data$Species <- forcats::fct_explicit_na(tax_data$Species, 'sp.')
```
```{r do the same for chlorplast data}
Chloroplast_tax_data$tax.Kingdom <- ifelse(Chloroplast_tax_data$boot.Kingdom <= mythreshold, NA, Chloroplast_tax_data$tax.Kingdom)
Chloroplast_tax_data$tax.Phylum <- ifelse(Chloroplast_tax_data$boot.Phylum <= mythreshold, NA, Chloroplast_tax_data$tax.Phylum)
Chloroplast_tax_data$tax.Class <- ifelse(Chloroplast_tax_data$boot.Class <= mythreshold, NA, Chloroplast_tax_data$tax.Class)
Chloroplast_tax_data$tax.Order <- ifelse(Chloroplast_tax_data$boot.Order <= mythreshold, NA, Chloroplast_tax_data$tax.Order)
Chloroplast_tax_data$tax.Family <- ifelse(Chloroplast_tax_data$boot.Family <= mythreshold, NA, Chloroplast_tax_data$tax.Family)
Chloroplast_tax_data$tax.Genus <- ifelse(Chloroplast_tax_data$boot.Genus <= mythreshold, NA, Chloroplast_tax_data$tax.Genus)
Chloroplast_tax_data$tax.Kingdom <- forcats::fct_explicit_na(Chloroplast_tax_data$tax.Kingdom, 'k_unassigned')
Chloroplast_tax_data$tax.Phylum <- forcats::fct_explicit_na(Chloroplast_tax_data$tax.Phylum, 'p_unassigned')
Chloroplast_tax_data$tax.Class <- forcats::fct_explicit_na(Chloroplast_tax_data$tax.Class, 'c_unassigned')
Chloroplast_tax_data$tax.Order <- forcats::fct_explicit_na(Chloroplast_tax_data$tax.Order, 'o_unassigned')
Chloroplast_tax_data$tax.Family <- forcats::fct_explicit_na(Chloroplast_tax_data$tax.Family, 'f_unassigned')
Chloroplast_tax_data$Genus <- forcats::fct_explicit_na(Chloroplast_tax_data$tax.Genus, 'g_unassigned')
Chloroplast_tax_data$Species <- forcats::fct_explicit_na(Chloroplast_tax_data$tax.Species, 'sp.')
```
```{r Select the columns you want and ditch the rest}
tax_data <- tax_data %>% select(ASV,tax.Kingdom,tax.Phylum,tax.Order,tax.Family,tax.Class,Genus,Species)
Chloroplast_tax_data <- Chloroplast_tax_data %>% rename(ASV = asv)
Chloroplast_tax_data <- Chloroplast_tax_data %>% select(ASV,tax.Kingdom,tax.Phylum,tax.Order,tax.Family,tax.Class, tax.Genus, tax.Species)
```
```{r Rename columns (I do this to be neat but you don't have to)}
tax_data <- tax_data %>% rename(Kingdom = tax.Kingdom)
tax_data <- tax_data %>% rename(Phylum = tax.Phylum)
tax_data <- tax_data %>% rename(Class = tax.Class)
tax_data <- tax_data %>% rename(Order = tax.Order)
tax_data <- tax_data %>% rename(Family = tax.Family)
Chloroplast_tax_data <- Chloroplast_tax_data %>% rename(Kingdom = tax.Kingdom)
Chloroplast_tax_data <- Chloroplast_tax_data %>% rename(Phylum = tax.Phylum)
Chloroplast_tax_data <- Chloroplast_tax_data %>% rename(Class = tax.Class)
Chloroplast_tax_data <- Chloroplast_tax_data %>% rename(Order = tax.Order)
Chloroplast_tax_data <- Chloroplast_tax_data %>% rename(Family = tax.Family)
Chloroplast_tax_data <- Chloroplast_tax_data %>% rename(Genus = tax.Genus)
Chloroplast_tax_data <- Chloroplast_tax_data %>% rename(Species = tax.Species)
```
```{r Remove Chloroplasts from Silva aka tax_data}
tax_data <- tax_data %>% filter(!Order %in% c('Chloroplast'))
```
```{r rbind the tax_data to Chloroplast_tax_data}
tax_data <- rbind(tax_data,Chloroplast_tax_data)
```
```{r Left join taxonomy data and ASV data to keep things simple}
tax_data$ASV <- as.character(tax_data$ASV) # Must be same type for join to work
asv_data$ASV <- as.character(asv_data$ASV) # Must be same type for join to work
asv_data <- left_join(asv_data, tax_data,by = c("ASV" = "ASV")) # Vdentifies cols with shared IDs
#print(asv_data)
```
```{r Make a long version of your data}
asv_data_long <- gather(data = asv_data, key = SampleID, value = Abundance, -c(1,92,93,94,95,96,97,98))
```
```{r Left join your sample_data}
asv_data_long <- asv_data_long %>% left_join(sample_data)
asv_data_long <- asv_data_long %>% select(!SampleID) %>% rename(SampleID = code)
```
```{r Remove unwanted ASVs - we will then store them in "Discarded_ASVs" in the case you want them later on.For this step we are going to rename our asv_data_long to Remove_1, Remove_2 and so on.. We will then label the final long ASV table as asv_cleaned_long.}
table(asv_data_long$Kingdom)
```
```{r Remove Archaea... 29602 ASVs total in the dataset (present >1 time ever)}
Remove_1 <- asv_data_long %>% filter(!Kingdom %in% c('Archaea'))
Archaea <- asv_data_long %>% filter(Kingdom %in% c('Archaea'))
```
```{r Remove Unassigned ASVs}
Remove_2 <- Remove_1 %>% filter(!Kingdom %in% c('k_unassigned'))
Unassigned_ASVs <- Remove_1 %>% filter(Kingdom %in% c('k_unassigned'))
```
```{r Remove Mitochondria - this is the last step of removals so rather than labelling "Remove_5" we are going for asv_clean_long}
asv_cleaned_long <- Remove_2 %>% filter(!Family %in% c('Mitochondria'))
Mitochondria <- Remove_2 %>% filter(Order %in% c("Mitochondria"))
```
```{r Make a table to check everything has been removed}
table(asv_cleaned_long$Kingdom)
```
```{r Make a dataframe for your stored Discarded ASVS "Discared_ASVs"}
Discarded_ASVS<-rbind(Archaea,Eukaryota,Chloroplast)
```
```{r Remove any unwanted samples. In this case I'm removing some weird samples that aren't mine}
asv_cleaned_long <- asv_cleaned_long %>% filter(!SampleID %in% c('L25','L26','L27','L28','L29','L30'))
```
```{r Quality filtration - theres a nature paper - basically you are removing anything that has a relative abundance of below 0.05%}
asv_cleaned_long <- asv_cleaned_long %>% arrange(desc(asv_cleaned_long$Abundance))
```
```{r Make your No_Filter table}
No_Filter <- asv_cleaned_long %>% filter(Abundance > 0)
```
```{r Calculate sample total column and left join back to No_Filter, this will now be called No_Filter_Total}
Sample_Total <- No_Filter %>% group_by(SampleID) %>% summarise(Total=sum(Abundance))
No_Filter_Total <- No_Filter %>% left_join(Sample_Total, c("SampleID"="SampleID"))
```
```{r We will then calculate the relative abundance and remove the bottom 0.05%. Once this is done we will relable it asv_cleaned_long as we are done}
No_Filter_Total$PercT <- ((No_Filter_Total$Abundance)/No_Filter_Total$Total)*100 #this makes it into a %
No_Filter_Total <- No_Filter_Total %>% arrange(desc(No_Filter_Total$PercT))
asv_cleaned_long <- No_Filter_Total %>% filter(PercT > 0.005)
```
```{r Store your filtered out ASVs incase you want to peak at them later on}
Filtered_ASVs_0.05 <- No_Filter_Total %>% filter(PercT < 0.005)
```
```{r Select the columns we want or aka delete those columns we created before. We cannot use that old relative abundance column, we must calculate a new RA based off of our now cleaned data set.}
asv_cleaned_long <- asv_cleaned_long %>% select(ASV,Kingdom,Phylum,Order,Family,Class,Genus,Species,SampleID,Abundance,Location,`Koppen`,`Lat DD`,Location_Rep,Type)
```
```{r Now we will name the ASVs}
asv_cleaned_long_name <- asv_cleaned_long %>% filter(!Abundance %in% c("0"))
asv_cleaned_long <- asv_cleaned_long_name %>% arrange(desc(asv_cleaned_long_name$Abundance))
asv_cleaned_by_asv <- asv_cleaned_long %>%group_by(ASV)
ASV_list <- asv_cleaned_by_asv %>% dplyr::select (ASV)
ASV_list <-distinct(ASV_list)
ASV_list$numbers<-seq(1000000001,1000000000+nrow(ASV_list),1)
ASV_list$prefix<-rep('B',nrow(ASV_list))
ASV_list <- ASV_list %>% unite(ASV_name, c('prefix','numbers'), sep='', remove=T)
asv_cleaned_long <- asv_cleaned_long %>% left_join(ASV_list , c("ASV"="ASV"))
asv_cleaned_long <- asv_cleaned_long %>% unite('FG', c(Family, Genus), remove=F, sep=';')
asv_cleaned_long <- asv_cleaned_long %>% unite('FGSp', c(FG, Species), remove=F, sep=';')
asv_cleaned_long <- asv_cleaned_long %>% unite('FGID', c(FGSp, ASV_name), remove=F, sep=';')
asv_cleaned_long <- asv_cleaned_long %>% unite('GSID', c(Genus, Species, ASV_name), remove=F, sep=';')
asv_cleaned_long <- asv_cleaned_long %>% unite('FGID.Location', c(FGID, Location), remove=F, sep=';')
asv_cleaned_long <- asv_cleaned_long %>% unite('taxonomy', c(Kingdom,Phylum,Class,Order,Family,Genus,Species), remove=F, sep=';')
```
```{r write file}
setwd("~/Dropbox/UTS/PhD/Projects/Chapter1_Vibrio_latitude_study/Data/16S/Data")
write_csv(asv_cleaned_long, "asv_cleaned_long_Chloroplasts.csv")
```