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
singularity: "docker://kkmann/gose-6mo-imputation@sha256:02a8f55c53d6f4917193abda823938e1ad7e8891fd6c392bd94f509e058eb34b"
configfile: "config.yml"
rule import_neurobot_csv:
output:
"data/{version}/df_baseline.rds",
"data/{version}/df_ctmri.rds",
"data/{version}/df_imaging.rds",
"data/{version}/df_labs.rds",
"data/{version}/df_gose.rds"
shell:
"""
Rscript scripts/import_neurobot_data.R data/{wildcards.version} data/{wildcards.version}
"""
rule prepare_data:
input:
rules.import_neurobot_csv.output,
markdown = "reports/prepare_data.Rmd"
output:
"output/{version}/data/df_gose.rds",
"output/{version}/data/df_baseline.rds",
"output/{version}/prepare_data.pdf",
figures = "output/{version}/prepare_data_figures.zip"
shell:
"""
mkdir -p output/{wildcards.version}/data
Rscript -e "rmarkdown::render('{input.markdown}', output_dir = 'output/{wildcards.version}', params = list(datapath = '../data/{wildcards.version}', max_lab_days = {config[max_lab_days]}, seed = {config[seed]}))"
mv reports/*.rds output/{wildcards.version}/data
mv reports/figures.zip {output.figures}
"""
rule impute_baseline:
input:
rules.prepare_data.output
output:
["output/{version}/data/mi_baseline/df_baseline_mi_%i.rds" % i for i in range(1, config["mi_m"] + 1)]
shell:
"""
mkdir -p output/{wildcards.version}/data/mi_baseline
Rscript scripts/impute_baseline.R output/{wildcards.version}/data/df_baseline.rds output/{wildcards.version}/data/mi_baseline {config[mi_m]} {config[mi_maxiter]} {config[seed]}
"""
rule generate_validation_data:
input:
rules.prepare_data.output,
rules.impute_baseline.output
output:
["output/{version}/data/validation/df_%s_mi_%i_fold_%i.rds" % (s, i, j)
for s in ("train", "test")
for i in range(1, config["mi_m"] + 1)
for j in range(1, config["folds"] + 1)
]
shell:
"""
mkdir -p output/{wildcards.version}/data/validation
Rscript scripts/generate_validation_data.R output/{wildcards.version}/data {config[mi_m]} {config[folds]} {config[seed]}
"""
# adjust threads by model type
def get_rule_threads(wildcards):
if wildcards.model in ("locf", "msm"):
return 1
else:
return config["stan"]["chains"]
rule fit_model_validation_set:
input:
"config.yml",
"output/{version}/data/validation/df_train_mi_{i}_fold_{j}.rds"
output:
"output/{version}/data/validation/posteriors/{model}/df_posterior_mi_{i}_fold_{j}.rds"
threads:
get_rule_threads
shell:
"""
mkdir -p output/{wildcards.version}/data/validation/posteriors/{wildcards.model}
Rscript models/{wildcards.model}/fit.R {input[1]} {output}
"""
# helper rule to just build all posterior datasets
rule model_posteriors:
input:
["output/v1.1/data/validation/posteriors/%s/df_posterior_mi_%i_fold_%i.rds" % (m, i, j)
for m in ("locf", "msm", "gp", "gp_nb", "mm", "mm_nb")
for i in range(1, config["mi_m"] + 1)
for j in range(1, config["folds"] + 1)
]
rule model_assessment:
input:
pop_report = rules.prepare_data.output,
posteriors = rules.model_posteriors.input,
markdown = "reports/model_assessment.Rmd"
output:
pdf = "output/{version}/model_assessment.pdf",
figures = "output/{version}/model_assessment_figures.zip"
shell:
"""
mkdir -p output/{wildcards.version}
Rscript -e "rmarkdown::render('{input.markdown}', output_dir = 'output/{wildcards.version}', params = list(data_dir = '../output/{wildcards.version}/data', config_file = '../config.yml'))"
mv reports/figures.zip {output.figures}
"""
rule generate_imputation_data:
input:
rules.prepare_data.output,
rules.impute_baseline.output
output:
["output/{version}/data/imputation/df_mi_%i.rds" % i
for i in range(1, config["mi_m"] + 1)
]
shell:
"""
mkdir -p output/{wildcards.version}/data/imputation
Rscript scripts/generate_imputation_data.R output/{wildcards.version}/data {config[mi_m]}
"""
# rules for imputing on entire dataset
rule model_impute:
input:
"config.yml",
"output/{version}/data/imputation/df_mi_{i}.rds"
output:
"output/{version}/data/imputation/{model}/df_gose_imputed_mi_{i}.rds"
threads:
get_rule_threads
shell:
"""
mkdir -p output/{wildcards.version}/data/imputation/{wildcards.model}
Rscript models/{wildcards.model}/fit.R {input[1]} {output}
"""
# final reported values are a combination of the imputed and per-protocol
# observed ones
rule post_process_imputations:
input:
"config.yml",
["output/{version}/data/imputation/{model}/df_gose_imputed_mi_%i.rds" % i
for i in range(1, config["mi_m"] + 1)
]
output:
"output/{version}/data/imputation/{model}/df_gose_imputed.csv"
shell:
"""
mkdir -p output/{wildcards.version}/data/imputation/{wildcards.model}
Rscript scripts/post_process_imputations.R output/{wildcards.version}/data/imputation/{wildcards.model} output/{wildcards.version}/data/df_gose.rds {output}
"""