sc_morph_tracking/10_slingshot.R at main · CarragherLab/sc_morph_tracking · 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
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

library(tidyverse)
library(data.table)
#library(ggridges)
library(viridis)
library(slingshot)
library(mclust)
library(ggpubr)
#library(RColorBrewer)


#########################################Gaussian G=5 clusters current best fit####################################################


df <- fread("../data/08_cleaned_umap.csv")


timecourse <- df %>%
  filter(Metadata_library=="timecourse")

timecourse$Metadata_plate <- as.factor(timecourse$Metadata_plate)

timecourse <- timecourse %>%
  mutate_at("Metadata_plate", as.factor)

timecourse <- timecourse[order(timecourse$Metadata_plate),]

timecourse_sub <- timecourse %>%
  group_by(Metadata_plate)%>%
  slice_sample(n=2000) %>%
  ungroup()

rm(df)


#get drugs

drugs_df <- df %>%
  filter(Metadata_library != "timecourse") %>%
  select(X1, X2, Metadata_compound, Metadata_concentration)

drugs_rd <- df %>%
  filter(Metadata_library != "timecourse") %>%
  select(X1, X2)


#for kmeans clusters

#plot clusters on umap for kmeans 1:12 clusters
for (i in 1:12){
  cl <- kmeans(rd, centers = i)$cluster
  timecourse_sub$cl <- as.factor(cl)
  p <- ggplot(timecourse_sub, aes(X1, X2))+
  geom_point(aes(colour = cl), alpha = 0.5)+
  #scale_colour_viridis_d(option="C")+
  labs(colour="cluster")
  print(p)

}

#plot clusters on umap for gaussian 5:12 clusters
for (i in 5:12){
  cl <- Mclust(rd, G = i)$classification
  timecourse_sub$cl <- as.factor(cl)
  p <- ggplot(timecourse_sub, aes(X1, X2))+
  geom_point(aes(colour = cl), alpha = 0.5)+
  #scale_colour_viridis_d(option="C")+
  labs(colour="cluster")
  print(p)

}

#create cluster labels
cl <- Mclust(rd, G = 5)$classification

#add cluster labels to dataframe
timecourse_sub$cl <- as.factor(cl)

ggplot(timecourse_sub, aes(X1, X2))+
  geom_point(aes(colour = cl), alpha = 0.5)+
  #scale_colour_viridis_d(option="C")+
  labs(colour="cluster")

fwrite(timecourse_sub, "../data/10_slingshot_clusters.csv")