-
Notifications
You must be signed in to change notification settings - Fork 0
/
Sym_densities_size_vol
249 lines (185 loc) · 8.42 KB
/
Sym_densities_size_vol
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
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
---
title: "Cell sizes, densities, and total volume symbiont : host"
author: "Kira Turnham"
date: '2022-12-16'
---
```{r}
library(tidyverse)
library(ggpubr)
library(dplyr)
library(ggplot2)
library(grid)
library(fBasics)
library(tidyr)
library(reshape2)
library(agricolae)
library(Rmisc)
library(gdata)
library(rlist)
library("ggsci")
library(MASS)
library (splines)
library(ggpubr)
#read in data files
cellsize=fread("sym_size.csv")
celldensity=fread("sym_cell_density.csv")
tot_vol=fread("sym_vol.csv")
#assign spp. colors
group.colors <- c("C.latusorum"= "#8b5e3c", "D.glynnii"= "#fbb040")
#cell size
d_sub<- subset(cellsize, sym.spp=="D.glynnii")
c_sub<-subset(cellsize, sym.spp=="C.latusorum")
#summarize by colony and spp
sum.dl<-summarySE(d_sub, measurevar="length", groupvars=c( "colony"))
sum.dw<-summarySE(d_sub, measurevar="width", groupvars=c( "colony"))
sum.cl<-summarySE(c_sub, measurevar="length", groupvars=c( "colony"))
sum.cw<-summarySE(c_sub, measurevar="width", groupvars=c( "colony"))
#merge summarized data back together
mergedlw <- cbind(sum.dl$colony, sum.dl$length, sum.dw$width)
mergeclw <- cbind(sum.cl$colony, sum.cl$length, sum.cw$width)
merged<- rbind(mergedlw, mergeclw)
p<-data_frame(merged)
data_new <- type.convert(p, as.is = TRUE)
d<-data_frame(mergedlw)
c<-data_frame(mergeclw)
#plot all measurements
csp<-ggplot(cellsize, aes(x=width,y=length, shape=sym.spp, color=sym.spp))+
geom_point(alpha=0.1)+
scale_color_manual(values=group.colors)
#Now overlay averages in bold
#had to make csv manually of the average cell sizes
all<-read.csv("all_sum_cell_size.csv")
COLS=c("#8b5e3c", "#fbb040")
names(COLS)=c("C.latusorum", "D.glynnii")
sum.width<-summarySE(cellsize, measurevar="width", groupvars=c("sym.spp"))
# sym.spp N width sd se ci
#1 C 531 10.07932 0.9630256 0.04179175 0.08209780
#2 D 421 8.76736 0.8956324 0.04365045 0.08580057
sum.length<-summarySE(cellsize, measurevar="length", groupvars=c("sym.spp"))
# sym.spp N length sd se ci
#1 C 531 11.015745 0.9166228 0.03977804 0.07814198
#2 D 421 9.683104 0.9058834 0.04415005 0.08678260
sum.size<-summarySE(cellsize, measurevar="length", groupvars=c("sym.spp"))
cellsize %>%
ggviolin( x = "sym.spp", y = "width", fill = "sym.spp",
palette = c(values=COLS),
add = "boxplot", add.params = list(fill = "white"))+
stat_compare_means(comparisons = sum.width, label = "p.signif") +
stat_compare_means(label.y = 1.5)
cellsize %>%
ggviolin( x = "sym.spp", y = "length", fill = "sym.spp",
palette = c(values=COLS),
add = "boxplot", add.params = list(fill = "white"))+
stat_compare_means(comparisons = sum.length, label = "p.signif") +
stat_compare_means(label.y = 1.5)
```
```{r setup, include=FALSE}
a<-csp + geom_point(all, mapping=aes(x=width,y=length, shape=sym.spp, color=sym.spp))+
geom_point(size=0.05)+
scale_shape_manual(values = c( 15, 0, 16, 1))+
scale_fill_manual(values=c("#48404d","lightsalmon"), name="Symbiont spp.",
labels = c(expression(italic("C. latusorum")), (expression(italic("D. glynnii")))))
#cs <- id %>%
# group_by(sym.spp) %>%
# summarise(colony = mean(colony))
#cell density per host tissue area
sum.density<- summarySE(celldensity, measurevar="cellspercm2", groupvars=c( "sym.spp"))
#barchart density
densitybar<-ggplot(data=sum.density, aes(x=sym.spp,y=cellspercm2, col=I("black")))+
geom_bar(position="dodge",stat='identity', fill=group.color)+
labs(x="Symbiont species",y='Cells/cm2 host tissue ', title = "Symbiont density")+
theme_classic()+
theme(legend.position="none",aspect.ratio = 2)+
geom_errorbar(aes(ymin=cellspercm2-sd, ymax=cellspercm2+sd), width=.2,
position=position_dodge(.9))
#boxplot density
den_bx<-ggplot(data=sum.density, aes(x=sym.spp,y=cellspercm2, fill=sym.spp))+
geom_boxplot()+
labs(x="Symbiont species",y='Symbiont size normalized to cell number', title = "Symbiont volume per host tissue")+
theme_classic()+
scale_fill_manual(values=group.color)
#calculate percent difference in C vs D densities
perc_diff<-sum.density %>%
summarise(cellspercm2 = (((last(cellspercm2) - first(cellspercm2)) /(((last(cellspercm2)+first(cellspercm2))/2))*100)))
#normality test
with(celldensity, shapiro.test(cellspercm2))#W = 0.94205, p-value = 0.5251
with(celldensity, shapiro.test(cellspercm2[sym.spp == "Clatusorum"])) #W = 0.9557, p-value = 0.786
with(celldensity, shapiro.test(cellspercm2[sym.spp == "Dglynnii"])) #W = 0.88203, p-value = 0.2785
#T-test
t.test(cellspercm2 ~ sym.spp, data=celldensity, alternative = "two.sided", var.equal = TRUE)
#t = -3.497, df = 10, p-value = 0.005755
#alternative hypothesis: true difference in means is not equal to 0
#95 percent confidence interval:
# -2084449.3 -461989.4
#sample estimates:
#mean in group C. latusorum mean in group D. glynnii
# 3095985 4369204
### sum symbiont volume to host cm2 tissue
sum.tot.vol<- summarySE(tot_vol, measurevar="divise", groupvars=c("sym.spp"))
#sym.spp N divise sd se ci
# C 8 7714.296 2426.500 857.8974 2028.605
# D 8 12039.736 4733.408 1673.5126 3957.228
#barplot for total volume of symbionts per host cm2 tissue (taking into account size and density)
barplotvol<-ggplot(data=sum.tot.vol, aes(x=sym.spp,y=divise, fill=sym.spp))+
geom_bar(stat='identity', fill=group.color)+
labs(x="Symbiont species",y='Total symbiont size normalized to cell number', title = "Symbiont volume per host tissue")+
theme_classic()+
theme(legend.position="none",aspect.ratio = 2)+
scale_fill_manual(values=group.colors)+
geom_errorbar(aes(ymin=divise-ci, ymax=divise+ci), width=.2,
position=position_dodge(.9))
#normality test for total symbiont volume per host area
with(tot_vol, shapiro.test(divise))#W = 0.91796, p-value = 0.1564
with(tot_vol, shapiro.test(divise[sym.spp == "C"])) #W = 0.84743, p-value = 0.08971
with(tot_vol, shapiro.test(divise[sym.spp == "D"])) #W = 0.90465, p-value = 0.3179
#T-test
t.test(divise ~ sym.spp, data=tot_vol, alternative = "two.sided", var.equal = FALSE)
#t = -2.3, df = 14, p-value = 0.03735
################
#cell volume calculation
d_sub$volume <- ((4/3)*pi*((0.5*d_sub$length)*(0.5*d_sub$width)*(0.5*d_sub$width)))
c_sub$volume <- ((4/3)*pi*((0.5*c_sub$length)*(0.5*c_sub$width)*(0.5*c_sub$width)))
d_avg<- mean(d_sub$volume)
#D. glynnii average cell volume 425.8577 um^2
summarySE(d_sub, measurevar("volume"))
c_avg<-mean(c_sub$volume)
#C. latusorum average cell volume = 594.7636 um^2
sum.c_vol<-summarySE(c_sub, measurevar="volume")
# .id N volume sd se ci
#1 <NA> 477 594.7636 158.7554 7.268909 14.28312
sum.d_vol<-summarySE(d_sub, measurevar="volume")
# .id N volume sd se ci
#1 <NA> 475 425.8577 138.5505 6.357134 12.49165
merged.vol<-rbind(d_sub, c_sub)
merged.sum.vol<-rbind(sum.c_vol, sum.d_vol)
merged.vol %>%
ggviolin( x = "sym.spp", y = "volume", fill = "sym.spp",
palette = c(values=COLS),
add = "boxplot", add.params = list(fill = "white"))+
stat_compare_means(comparisons = merged.sum.vol, label = "p.signif") +
stat_compare_means(label.y = 1500)
vol_bx<-ggplot(data=merged.vol, aes(x=sym.spp,y=volume, fill=sym.spp))+
geom_boxplot()+
labs(x="Symbiont species",y='Symbiont size normalized to cell number', title = "Symbiont cell volume")+
theme_classic()+
scale_fill_manual(values=group.colors)
#calculate percent difference in sizes
perc_diff_vol<-merged.sum.vol %>%
summarise(volume = (((last(volume) - first(volume)) /(((last(volume)+first(volume))/2))*100)))
# -33.09863
# test normality
with(merged.vol, shapiro.test(volume))#W = 0.96847, p-value = 1.678e-13
with(merged.vol, shapiro.test(volume[sym.spp == "C"])) #W = 0.96561, p-value = 3.94e-09
with(merged.vol, shapiro.test(volume[sym.spp == "D"])) #W = 0.94496, p-value = 2.814e-12
#non normal, so need to use non-parametric test. These are independent samples, so Mann Witney U Test (i.e. wilcoxen but paired=FALSE)
wilcox.test(merged.vol$volume~
merged.vol$sym.spp)
#W = 182400, p-value < 2.2e-16
#alternative hypothesis: true location shift is not equal to 0
%>%
ggviolin( x = "sym.spp", y = "length", fill = "sym.spp",
palette = c(values=COLS),
add = "boxplot", add.params = list(fill = "white"))+
stat_compare_means(comparisons = sum.length, label = "p.signif") +
stat_compare_means(label.y = 1.5)
```