PopGen/ColonyPlates.R at master · wrshoemaker/PopGen · GitHub

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
library(xlsx)
library(vegan)
library(reshape)
library(ggplot2)
library(Rmisc)
mydata <- read.xlsx("~/Box Sync/LabPhotos/JanthinoProjectKJW/JanthinoProjectKJW_Colony_Counts.xlsx", 1)
sapply(mydata, class)

mydata$White <- as.numeric(as.character(mydata$White))
mydata$Purple.wr <- as.numeric(as.character(mydata$Purple.wr))
mydata$Purple.S <- as.numeric(as.character(mydata$Purple.S))

sapply(mydata, class)
# Add column of means

mydata.noNaN <- mydata[is.finite(mydata$White) & is.finite(mydata$Purple.wr) & is.finite(mydata$Purple.S), ]
# Add column with total colony count
mydata.noNaN$Total <- rowSums(mydata.noNaN[,6:8])
mydata.noNaN$H <- with(mydata.noNaN, diversity(mydata.noNaN[,6:8]))
mydata.noNaN$TreatDateSampDilDays <- do.call(paste,
                                             c(mydata.noNaN[c("Treatment", "SamplingDate", "PlateDilution", "PictureDate.days.since.plating.")],
                                               sep = ""))


mydata.Day7 <- mydata.noNaN[ which(mydata.noNaN$PictureDate.days.since.plating==7),]

# We don't trust plates with less than 30 colonies.
# Let's remove them.
mydata.Day7.subset <- subset(mydata.Day7, sum(mydata.Day7[,6:8]) >= 30)

mydata.Day7.Dil6 <- mydata.Day7[ which(mydata.Day7.subset$PlateDilution==-6),]

mydata.Day7.Dil6.Total.SE <- summarySE(mydata.Day7.Dil6, measurevar="Total", groupvars=c("TreatDateSampDilDays", "Treatment", "SamplingDate"),
                                    conf.interval = 0.95, .drop = TRUE)

mydata.Day7.Dil6.H.SE <- summarySE(mydata.Day7.Dil6, measurevar="H", groupvars=c("TreatDateSampDilDays", "Treatment", "SamplingDate"),
                        conf.interval = 0.95, .drop = TRUE)
mydata.Day7.Dil6.White.SE <- summarySE(mydata.Day7.Dil6, measurevar="White", groupvars=c("TreatDateSampDilDays", "Treatment", "SamplingDate"),
                                       conf.interval = 0.95, .drop = TRUE)

mydata.Day7.Dil6.Purple.S.SE <- summarySE(mydata.Day7.Dil6, measurevar="Purple.S", groupvars=c("TreatDateSampDilDays", "Treatment", "SamplingDate"),
                                       conf.interval = 0.95, .drop = TRUE)

mydata.Day7.Dil6.Purple.wr.SE <- summarySE(mydata.Day7.Dil6, measurevar="Purple.wr", groupvars=c("TreatDateSampDilDays", "Treatment", "SamplingDate"),
                                          conf.interval = 0.95, .drop = TRUE)


# Reshape data into long form for stacked bar chart
mydata.Day7.long <- melt(subset(mydata.Day7, select = -c(H,Total)),
            id.var= c("Treatment", "SamplingDate", "PlateDilution", "PictureDate.days.since.plating.", "TreatDateSampDilDays", "Replicate"))

# plot stacked bar chart for each plate dilution for 7-day old plates
ggplot(subset(mydata.Day7.long, PlateDilution == -5), aes(x=SamplingDate, y=value, fill=variable)) +
  geom_bar(stat="identity") + facet_grid(~Treatment) + theme_bw()

ggplot(subset(mydata.Day7.long, PlateDilution == -6), aes(x=SamplingDate, y=value, fill=variable)) +
  geom_bar(stat="identity") + facet_grid(~Treatment) + theme_bw()

ggplot(subset(mydata.Day7.long, PlateDilution == -7), aes(x=SamplingDate, y=value, fill=variable)) +
  geom_bar(stat="identity") + facet_grid(~Treatment) + theme_bw()


# We see here that what dilution is appropriate depends on the sampling date
# This is a technical issue.
#### To do: For each sample, pick an appropriate dilution. Normal <30 & >300 threshold isn't appropriate, since
#### we'll loose a lot of data that was countable


# Now plot Shannon's diversity

ggplot(data=mydata.Day7.Dil6.H.SE, aes(x=SamplingDate, y=H, fill=Treatment)) +
  geom_bar(stat="identity", position=position_dodge(), colour="black")