-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathgenerateFigures_model.jl
184 lines (161 loc) · 8.29 KB
/
generateFigures_model.jl
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
#push!(LOAD_PATH, "Utils/")
using Statistics, StatsBase
using Distributions
using Gadfly, StatsPlots
using Cairo, Fontconfig
include("utils.jl")
###### Global Var ####
figure_prefix = "_generated_figures/model/"
datasets= ["gCSI"]#["gray", "gCSI", "ctrpv2"]
expId_list = mapreduce(dt -> getExpId_h5(dt), vcat, datasets);
nId_list = map(dt -> length(getExpId_h5(dt)), datasets);
si = StrIndex(expId_list)
###### Data ##########
data_df = mapreduce(dt -> getRawData_h5(dt, false, si), vcat, datasets);
ml_df = mapreduce(dt -> getMLestimates(dt, si), vcat, datasets);
### example experiments from gCSI
expId_subset_list = ["NCI-H1648_AZ-628_8h", "Calu-1_PF-4708671_6b", "RERF-LC-MS_Gemcitabine_4b", "HCC78_Lapatinib_11a"];
data_subset = filter(:exp_id => x -> si.id2str[x] ∈ expId_subset_list, data_df);
ml_subset = filter(:exp_id => x -> si.id2str[x] ∈ expId_subset_list, ml_df);
posterior_subset = getPosterior_h5("gCSI", false, si, expId_subset_list);
### Raw data overlook
maxConcentration_df = combine(groupby(data_df, :exp_id),
:Concentration => maximum => :Concentration,
[:Concentration, :Viability] => ((a,b) -> b[argmax(a)]) => :Viability,
[:Concentration, :dataset] => ((a,b) -> b[argmax(a)]) => :dataset);
minConcentration_df = combine(groupby(data_df, :exp_id),
:Concentration => minimum => :Concentration,
[:Concentration, :Viability] => ((a,b) -> b[argmin(a)]) => :Viability,
[:Concentration, :dataset] => ((a,b) -> b[argmin(a)]) => :dataset)
## Draw Viability of max [] distributions per dataset
Gadfly.set_default_plot_size(4inch, 3inch)
x = percentile(maxConcentration_df.Viability, [2.5, 97.5])
ymin = [0.,0.]
ymax = ymin .+ 0.20
p1 = Gadfly.plot(maxConcentration_df, x=:Viability, color=:dataset,
Geom.histogram(position=:stack, bincount=200, density=true),
layer(x=x, ymin=ymin, ymax=ymax, Geom.ribbon()),
Coord.cartesian(xmin=-20, xmax=150),
Theme(panel_stroke="black"))
draw(PDF(figure_prefix*"viability_max_conc_LARGE.pdf", 4inch, 3inch), p1)
## Draw Viability of min [] distributions per dataset
Gadfly.set_default_plot_size(4inch, 3inch)
x = percentile(minConcentration_df.Viability, [2.5, 97.5])
ymin = [0.,0.]
ymax = ymin .+ 0.20
p2 = Gadfly.plot(minConcentration_df, x=:Viability, color=:dataset,
Geom.histogram(position=:stack, bincount=200, density=true),
layer(x=x, ymin=ymin, ymax=ymax, Geom.ribbon()),
Coord.cartesian(xmin=0, xmax=150),
Theme(panel_stroke="black"))
draw(PDF(figure_prefix*"viability_min_conc_LARGE.pdf", 4inch, 3inch), p2)
## Draw [] distributions per dataset
Gadfly.set_default_plot_size(4inch, 3inch)
x = percentile(data_df.Concentration, [2.5, 97.5])
ymin = [0.,0.]
ymax = ymin .+ 2.
p3 = Gadfly.plot(data_df, x=:Concentration, color=:dataset,
Geom.histogram(position=:stack, density=true),
layer(x=x, ymin=ymin, ymax=ymax, Geom.ribbon()),
Coord.cartesian(xmin=-7),
Theme(panel_stroke="black"))
draw(PDF(figure_prefix*"concentration_LARGE.pdf", 4inch, 3inch), p3)
### Prior density plots
N = 400000
## IC50
ic50_prior = rand(Normal(0,10), N)
x = percentile(data_df.Concentration, [2.5, 97.5])
ymin = [0.,0.]
ymax = ymin .+ 0.04
p4 = Gadfly.plot(layer(x=ic50_prior, Geom.density(bandwidth=1)),
layer(x=x, ymin=ymin, ymax=ymax, Geom.ribbon(),
Theme(panel_stroke="black")))
draw(PDF(figure_prefix*"ic50_prior_LARGE.pdf", 4inch, 3inch), p4)
## LDR
ldr_prior = rand(Normal(100,10), N)
x = percentile(minConcentration_df.Viability, [2.5, 97.5])
ymin = [0.,0.]
ymax = ymin .+ 0.04
p5 = Gadfly.plot(layer(x=ldr_prior, Geom.density(bandwidth=1)),
layer(x=x, ymin=ymin, ymax=ymax, Geom.ribbon(),
))
draw(PDF(figure_prefix*"ldr_prior.pdf", 4inch, 3inch), p5)
## HDR
λₖ = [0.4, 0.5, 0.1]
hdr_mm = MixtureModel([SkewNormal(0, 10, 1), Uniform(0, 100), SkewNormal(100, 20, -5)], λₖ)
hdr_prior = rand(hdr_mm, N)
x = percentile(maxConcentration_df.Viability, [2.5, 97.5])
ymin = [0.,0.]
ymax = ymin .+ 0.04
p6 = Gadfly.plot(layer(x=hdr_prior, Geom.density(bandwidth=1)),
layer(x=x, ymin=ymin, ymax=ymax, Geom.ribbon() ))
draw(PDF(figure_prefix*"hdr_prior.pdf", 4inch, 3inch), p6)
## slope
slope_prior = rand(LogNormal(0.5, 1), N)
p7 = Gadfly.plot(layer(x=slope_prior, Geom.density(bandwidth=1)),
Coord.cartesian(xmin=-5, xmax=40))
draw(PDF(figure_prefix*"slope_prior.pdf", 4inch, 3inch), p7)
## σ
σ_prior = rand(LogNormal(1, 1), N)
p8 = Gadfly.plot(layer(x=σ_prior, Geom.density(bandwidth=1)),
Coord.cartesian(xmin=-5, xmax=50))
draw(PDF(figure_prefix*"sigma_prior.pdf", 4inch, 3inch), p8)
### Examples of analysis
Gadfly.set_default_plot_size(6inch, 4inch)
p9 = Gadfly.plot(layer(data_subset, x=:Concentration, y=:Viability, color=si.id2str[data_subset.exp_id], Geom.point()), Scale.color_discrete())
for e in expId_subset_list
println(e)
postCurves = getPosteriorCurves(filter(:exp_id => x -> si.id2str[x] == e, posterior_subset), -4, 1)
xDose = parse.(Float64, names(postCurves))
med_curve = median.(eachcol(postCurves))
upper_curve = percentile.(eachcol(postCurves), 97.5)
lower_curve = percentile.(eachcol(postCurves), 2.5)
push!(p9, layer(x=xDose, y=med_curve, color=[e], linestyle=[:dot], Geom.line()))
mleFit = llogistic(Array(filter(:exp_id => x -> si.id2str[x] == e, ml_subset)[1, [:LDR, :HDR, :ic50, :slope]]))
yViability = mleFit.(xDose)
push!(p9, layer(x=xDose, y=yViability, color=[e], Geom.line()))
push!(p9, layer(x=xDose, ymin=lower_curve, ymax=upper_curve, color=[e], alpha=[0.3], Geom.ribbon()))
end
display(p9)
draw(PDF(figure_prefix*"curves_examples.pdf", 6inch, 4inch), p9)
### posterior distributions
### Add ΔHDR
Gadfly.set_default_plot_size(4inch, 2inch)
for e in expId_subset_list
tmp = filter(:exp_id => x -> si.id2str[x] == e, posterior_subset)
est = filter(:exp_id => x -> si.id2str[x] == e, ml_subset)[1, :HDR]
ΔHDR = diff(percentile(tmp.HDR, [2.5, 97.5]))[1]
p10 = Gadfly.plot(layer(x=hdr_prior, Geom.density(bandwidth=1)),
layer(tmp, x=:HDR, Geom.histogram(position=:stack, bincount=200, density=true)),
layer(xintercept=[est], Geom.vline(color=colorant"black")),
layer(xintercept=percentile(tmp.HDR, [2.5, 97.5]), Geom.vline()),
Coord.Cartesian(xmin=-10, xmax=110),
Guide.title("ΔHDR = $ΔHDR"))
draw(PDF(figure_prefix*"HDR_posterior_prior_"*e*".pdf", 4inch, 2inch), p10)
#display(p10)
end
for e in expId_subset_list
tmp = filter(:exp_id => x -> si.id2str[x] == e, posterior_subset)
est = filter(:exp_id => x -> si.id2str[x] == e, ml_subset)[1, :LDR]
p10 = Gadfly.plot(layer(x=ldr_prior, Geom.density(bandwidth=1)),
layer(tmp, x=:LDR, Geom.histogram(position=:stack, bincount=200, density=true)),
layer(xintercept=[est], Geom.vline(color=colorant"black")),
layer(xintercept=percentile(tmp.LDR, [2.5, 97.5]), Geom.vline()),
Coord.Cartesian(xmin=90, xmax=120))
draw(PDF(figure_prefix*"LDR_posterior_prior_"*e*".pdf", 4inch, 2inch), p10)
#display(p10)
end
### Add exp.dose-range
for e in expId_subset_list
tmp = filter(:exp_id => x -> si.id2str[x] == e, posterior_subset)
est = filter(:exp_id => x -> si.id2str[x] == e, ml_subset)[1, :ic50]
data_subset = filter(:exp_id => x -> x == si.str2id[e], data_df)
p10 = Gadfly.plot(layer(x=ic50_prior, Geom.density(bandwidth=1)),
layer(tmp, x=:ic50, Geom.histogram(position=:stack, bincount=200, density=true)),
layer(xintercept=[est], Geom.vline(color=colorant"black")),
layer(xintercept=percentile(tmp.ic50, [2.5, 97.5]), Geom.vline()),
layer(xintercept=[minimum(data_subset.Concentration), maximum(data_subset.Concentration)], Geom.vline(color=colorant"orange")),
Coord.Cartesian(xmin=-5, xmax=5))
draw(PDF(figure_prefix*"ic50_posterior_prior_"*e*".pdf", 4inch, 2inch), p10)
display(p10)
end