-
Notifications
You must be signed in to change notification settings - Fork 0
/
diseasy.pretty.stats.R
141 lines (106 loc) · 6.2 KB
/
diseasy.pretty.stats.R
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
cap2 = read.csv("H2Zcluster2")
as.data.frame(cap2)
caporg = read.csv("H2ZRanks")
as.data.frame(caporg)
fin = data.frame(caporg$gene, caporg$rankingSem, caporg$rankingSem, cap2$captureSemOpt, caporg$rankingTxt, caporg$captureTxt, cap2$captureTxtOpt)
semorthscores = vector()
for(htarget in hgenes[-1]){
b = orths[which(orths == htarget,arr.ind = TRUE)[1],-1]
for(zf in b){
if(zf != "" && !is.na(zf))
{semorthscores = append(semorthscores , as.numeric(hzsl[which(zgenes== zf), htarget]))
print(semorthscores)}
}
}
bs = 0
txtorthscores = vector()
for(htarget in hgenes[-1]){
b = orths[which(orths == htarget,arr.ind = TRUE)[1],-1]
for(zf in b){
if(zf != "" && !is.na(zf))
{txtorthscores = append( txtorthscores , as.numeric(hztl[which(zgenes== zf), htarget]))
print(txtorthscores)
bs = bs + 1}
}
}
for (z in zgenes) {
print(which(orths == z,arr.ind = TRUE)[1])
bs = bs + 1
}
mean(semorthscores)
sd(semorthscores)
mean(txtorthscores)
sd(txtorthscores)
mean(colMeans(as.data.frame(hzsl[,-1])))
sd(colMeans(as.data.frame(hzsl[,-1])))
mean(colMeans(as.data.frame(hztl[,-1])))
sd(colMeans(as.data.frame(hztl[,-1])))
aveScores = matrix(0,nrow = 3, ncol=4, dimnames = list(c("mn","sd","comp"),c("text", "textOrth","semantic","semanticOrth")))
aveScores[3,]=c("text", "textOrth","semantic","semanticOrth")
aveScores[1,1] =as.numeric(mean(colMeans(as.data.frame(hztl[,-1]))))
aveScores[2,1] =as.numeric(sd(colMeans(as.data.frame(hztl[,-1]))))
aveScores[1,2] = as.numeric(mean(txtorthscores))
aveScores[2,2] = as.numeric(sd(txtorthscores))
aveScores[1,3] = as.numeric(mean(colMeans(as.data.frame(hzsl[,-1]))))
aveScores[2,3] = as.numeric(sd(colMeans(as.data.frame(hzsl[,-1]))))
aveScores[1,4] = as.numeric(mean(semorthscores))
aveScores[2,4] = as.numeric(sd(semorthscores))
aveScores2 = as.data.frame(t(aveScores))
aveScores2[,1] = as.numeric(aveScores2[,1])
aveScores2[,2] = as.numeric(aveScores2[,2])
library(ggplot2)
ggplot(aveScores2, aes(x= comp, y=mn, fill=comp))+
geom_bar(stat = "identity")+
geom_errorbar(aes(ymin=mn-sd, ymax=mn+sd))+
scale_fill_brewer(palette="Set3")+
labs(x="Data", y="Mean", title = "Average Comparison Score vs Average Ortholog Score")
ggplot(fin, aes(x=caporg.rankingSem))+
geom_histogram(binwidth = 15, fill = "orange", color="black")+
labs(x="Rank", y="Counts", title = "Distribution of Ortholog Ranking- Semantic Comparison")+
geom_vline(xintercept=mean(fin$caporg.rankingSem),size=2, show.legend = TRUE)+
geom_vline(xintercept=median(fin$caporg.rankingSem),size=2, show.legend = TRUE, color="brown")
ggplot(fin, aes(x=caporg.rankingTxt))+
geom_histogram(binwidth = 15, fill = "cornflowerblue", color="black")+
labs(x="Rank", y="Counts", title = "Distribution of Ortholog Ranking- Textual Comparison")+
geom_vline(xintercept=mean(fin$caporg.rankingTxt),size=2, show.legend = TRUE)
ggplot(fin, aes(x=caporg.rankingSem.1))+
geom_histogram(binwidth = 15, fill = "orange", color="black")+
labs(x="Rank capturing ortholog in cluster", y="Counts", title = "Distribution of Ortholog Ranking- Semantic Comparison, Cluster distance <0.1")+
geom_vline(xintercept=mean(fin$caporg.rankingSem.1),size=2, show.legend = TRUE)
ggplot(fin, aes(x=caporg.captureTxt))+
geom_histogram(binwidth = 15, fill = "cornflowerblue", color="black")+
labs(x="Rank capturing ortholog in cluster", y="Counts", title = "Distribution of Ortholog Ranking- Textual Comparison, Cluster distance <0.1")+
geom_vline(xintercept=mean(fin$caporg.captureTxt),size=2, show.legend = TRUE)
ggplot(fin, aes(x=cap2.captureSemOpt))+
geom_histogram(binwidth = 15, fill = "orange", color="black")+
labs(x="Rank capturing ortholog in cluster", y="Counts", title = "Distribution of Ortholog Ranking- Semantic Comparison, Optimal clusters")+
geom_vline(xintercept=mean(fin$cap2.captureSemOpt),size=2, show.legend = TRUE)
ggplot(fin, aes(x=cap2.captureTxtOpt))+
geom_histogram(binwidth = 15, fill = "cornflowerblue", color="black")+
labs(x="Rank capturing ortholog in cluster", y="Counts", title = "Distribution of Ortholog Ranking- Textual Comparison, Optimal clusters")+
geom_vline(xintercept=mean(fin$cap2.captureTxtOpt),size=2, show.legend = TRUE)
averanks = matrix(0, ncol = 4, nrow = 6)
averanks[1,] = c(mean(fin$caporg.rankingSem),sd(fin$caporg.rankingSem),"Rank Only","Semantic")
averanks[2,] = c(mean(fin$caporg.rankingTxt),sd(fin$caporg.rankingTxt),"Rank Only","Text")
averanks[3,] = c(mean(fin$caporg.rankingSem.1),sd(fin$caporg.rankingSem.1),"Cluster < 0.1","Semantic")
averanks[4,] = c(mean(fin$caporg.captureTxt),sd(fin$caporg.captureTxt),"Cluster < 0.1","Text")
averanks[5,] = c(mean(fin$cap2.captureSemOpt),sd(fin$cap2.captureSemOpt),"Optimal Cluster","Semantic")
averanks[6,] = c(mean(fin$cap2.captureTxtOpt),sd(fin$cap2.captureTxtOpt),"Optimal Cluster","Text")
colnames(averanks) = c("Median","Stand.Deviation","Method","Comparison")
averanks = as.data.frame(averanks)
averanks[,1]=as.numeric(averanks[,1])
averanks[,2]=as.numeric(averanks[,2])
ggplot(averanks, aes(x= reorder(Method, -Median), y=Median, fill=Comparison))+
geom_bar(stat = "identity", position = "dodge")+
scale_fill_brewer(palette="Set3")+
labs(x="Method", y="Median Ranking", title = "Median Rank by Clustering Method")
tsum.test(mean.x = averanks[1,1], s.x = averanks[1,2], n.x = 731, mean.y = 366, s.y = 277, n.y=731)
tsum.test(mean.x = averanks[2,1], s.x = averanks[2,2], n.x = 731, mean.y = 366, s.y = 277, n.y=731)
tsum.test(mean.x = averanks[3,1], s.x = averanks[3,2], n.x = 731, mean.y = 366, s.y = 277, n.y=731)
tsum.test(mean.x = averanks[4,1], s.x = averanks[4,2], n.x = 731, mean.y = 366, s.y = 277, n.y=731)
tsum.test(mean.x = averanks[5,1], s.x = averanks[5,2], n.x = 731, mean.y = 366, s.y = 277, n.y=731)
tsum.test(mean.x = averanks[6,1], s.x = averanks[6,2], n.x = 731, mean.y = 366, s.y = 277, n.y=731)
tsum.test(mean.x = mean(semorthscores), s.x = sd(semorthscores), n.x = 731, mean.y = mean(colMeans(as.data.frame(hzsl[,-1]))), s.y = sd(colMeans(as.data.frame(hzsl[,-1]))), n.y=731)
tsum.test(mean.x = mean(txtorthscores), s.x = sd(txtorthscores), n.x = 731, mean.y = mean(colMeans(as.data.frame(hztl[,-1]))), s.y = sd(colMeans(as.data.frame(hztl[,-1]))), n.y=731)
library(MASS)
fit <- fitdistr(fin$caporg.rankingSem,"normal")