-
Notifications
You must be signed in to change notification settings - Fork 14
/
14R-ML-fungi-cancer-stage.R
executable file
·272 lines (230 loc) · 19 KB
/
14R-ML-fungi-cancer-stage.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
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
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
#-----------------------------------------------------------------------------
# 14R-ML-fungi-cancer-stage.R
# Copyright (c) 2021--, Greg Poore
# Purposes:
# - Test whether fungi can predict stage I vs stage IV cancers
# - Test whether blood-derived fungi still distinguish cancer types in stages I-II
#-----------------------------------------------------------------------------
#----------------------------------------------------------#
# Load environments
#----------------------------------------------------------#
# Load dependencies
require(devtools)
require(doMC)
require(phyloseq)
require(microbiome)
require(vegan)
require(plyr)
require(dplyr)
require(reshape2)
require(ggpubr)
require(ggsci)
require(ANCOMBC)
require(biomformat)
require(Rhdf5lib)
numCores <- detectCores()
registerDoMC(cores=numCores)
#----------------------------------------------------------#
# Load TCGA fungi data
#----------------------------------------------------------#
load("Interim_data/data_for_pvca_tcga_taxa_levels_decontamV2_2Apr22.RData", verbose = T)
#----------------------------------------------------------#
# Add stage labels
#----------------------------------------------------------#
# Remove unclear or non-useful stages
metaQiitaCombined_Nonzero_DecontamV2_Path <- droplevels(metaQiitaCombined_Nonzero_DecontamV2[! (metaQiitaCombined_Nonzero_DecontamV2$pathologic_stage_label == "Not available" |
metaQiitaCombined_Nonzero_DecontamV2$pathologic_stage_label == "I or II NOS" |
metaQiitaCombined_Nonzero_DecontamV2$pathologic_stage_label == "Stage 0" |
metaQiitaCombined_Nonzero_DecontamV2$pathologic_stage_label == "Stage IS" |
metaQiitaCombined_Nonzero_DecontamV2$pathologic_stage_label == "Stage Tis" |
metaQiitaCombined_Nonzero_DecontamV2$pathologic_stage_label == "Stage X"),])
tumorStageVector <- factor(metaQiitaCombined_Nonzero_DecontamV2_Path$pathologic_stage_label)
levels(tumorStageVector) <- list(StageI = c("Stage I", "Stage IA", "Stage IB", "Stage IC"),
StageII = c("Stage II", "Stage IIA", "Stage IIB", "Stage IIC"),
StageIII = c("Stage III", "Stage IIIA", "Stage IIIB", "Stage IIIC"),
StageIV = c("Stage IV", "Stage IVA", "Stage IVB", "Stage IVC"))
metaQiitaCombined_Nonzero_DecontamV2_Path$pathologic_stage_label_binned <- tumorStageVector
table(metaQiitaCombined_Nonzero_DecontamV2_Path$pathologic_stage_label_binned)
#---------------------------Subset VSNM data to primary tumors---------------------------#
metaQiitaCombined_Nonzero_DecontamV2_Path_PT <- metaQiitaCombined_Nonzero_DecontamV2_Path %>%
filter(sample_type == "Primary Tumor") %>% droplevels()
rep200FungiDecontamV2SpeciesVSNM <- rep200FungiDecontamV2SpeciesVSNM_Obj$snmData
rep200FungiDecontamV2SpeciesVSNM_Path_PT <- rep200FungiDecontamV2SpeciesVSNM[rownames(metaQiitaCombined_Nonzero_DecontamV2_Path_PT),]
save(metaQiitaCombined_Nonzero_DecontamV2_Path_PT,
rep200FungiDecontamV2SpeciesVSNM_Path_PT,
file = "Interim_data/data_for_ml_stage_5Apr22.RData")
# See scripts: S23R
#---------------------------Subset VSNM data blood derived normals---------------------------#
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN <- metaQiitaCombined_Nonzero_DecontamV2_Path %>%
filter(sample_type == "Blood Derived Normal") %>%
filter(pathologic_stage_label_binned %in% c("StageI","StageII")) %>% droplevels()
rep200FungiDecontamV2SpeciesVSNM_Path_BDN <- rep200FungiDecontamV2SpeciesVSNM[rownames(metaQiitaCombined_Nonzero_DecontamV2_Path_BDN),]
save(metaQiitaCombined_Nonzero_DecontamV2_Path_BDN,
rep200FungiDecontamV2SpeciesVSNM_Path_BDN,
file = "Interim_data/data_for_ml_vsnm_bdn_stageI_II_5Apr22.RData")
# See scripts: S24R
#---------------------------Subset raw data per seq center to blood derived normals---------------------------#
load("Interim_data/data_for_ml_tcga_by_seq_center_and_experimental_strategy_with_coverage_filter_decontamV2_2Apr22.RData")
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_HMS <- metaQiitaCombined_Nonzero_DecontamV2_Path_BDN %>%
filter(data_submitting_center_label == "Harvard Medical School") %>% droplevels()
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_BCM <- metaQiitaCombined_Nonzero_DecontamV2_Path_BDN %>%
filter(data_submitting_center_label == "Baylor College of Medicine") %>% droplevels()
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_MDA <- metaQiitaCombined_Nonzero_DecontamV2_Path_BDN %>%
filter(data_submitting_center_label == "MD Anderson - Institute for Applied Cancer Science") %>% droplevels()
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_WashU <- metaQiitaCombined_Nonzero_DecontamV2_Path_BDN %>%
filter(data_submitting_center_label == "Washington University School of Medicine") %>% droplevels()
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_Broad_WGS <- metaQiitaCombined_Nonzero_DecontamV2_Path_BDN %>%
filter(data_submitting_center_label == "Broad Institute of MIT and Harvard") %>% droplevels()
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_HMS <- rep200_HiSeq_Fungi_DecontamV2_HMS[rownames(metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_HMS),]
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_BCM <- rep200_HiSeq_Fungi_DecontamV2_BCM[rownames(metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_BCM),]
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_MDA <- rep200_HiSeq_Fungi_DecontamV2_MDA[rownames(metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_MDA),]
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_WashU <- rep200_HiSeq_Fungi_DecontamV2_WashU[rownames(metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_WashU),]
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_Broad_WGS <- rep200_HiSeq_Fungi_DecontamV2_Broad_WGS[rownames(metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_Broad_WGS),]
save(rep200_HiSeq_Fungi_DecontamV2_Path_BDN_HMS,
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_BCM,
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_MDA,
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_WashU,
rep200_HiSeq_Fungi_DecontamV2_Path_BDN_Broad_WGS,
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_HMS,
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_BCM,
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_MDA,
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_WashU,
metaQiitaCombined_Nonzero_DecontamV2_Path_BDN_Broad_WGS,
file = "Interim_data/data_for_ml_raw_counts_bdn_stageI_II_5Apr22.RData")
# See scripts: S25R
#----------------------------------------------------------#
# Plot results for primary tumors
#----------------------------------------------------------#
source("Supporting_scripts/S00-SummarySE.R") # Contains a function that calculates std error and 95% confidence intervals
mlPerfAll10k_stage <- read.csv("Interim_data/rep_perfFungi_10k_rep1_tcga_stage_ALL_DecontamV2_5Apr22.csv", stringsAsFactors = FALSE)
abbreviationsTCGA_Allcancer <- read.csv("Supporting_data/tcga_abbreviations.csv", stringsAsFactors = FALSE, row.names = 1)
mlPerfAll10k_stage$abbrev <- abbreviationsTCGA_Allcancer[mlPerfAll10k_stage$diseaseType,"abbrev"]
mlPerfAll10k_stage <- mlPerfAll10k_stage[,!(colnames(mlPerfAll10k_stage) == "X")]
colnames(mlPerfAll10k_stage)[1:2] <- c("AUROC","AUPR")
# Add null perf values. Note: AUPR null is prevalence of **positive class**
# For stage, "StageIV" is the positive class and is used to calculate null AUPR
mlPerfAll10k_stage$nullAUPR <- ifelse(mlPerfAll10k_stage$minorityClassName == "StageIV",
yes=mlPerfAll10k_stage$minorityClassSize/(mlPerfAll10k_stage$minorityClassSize+mlPerfAll10k_stage$majorityClassSize),
no=mlPerfAll10k_stage$majorityClassSize/(mlPerfAll10k_stage$minorityClassSize+mlPerfAll10k_stage$majorityClassSize))
mlPerfAll10k_stage$nullAUROC <- 0.5
# Rename entries in the "datasetName" column
table(mlPerfAll10k_stage$datasetName)
mlPerfAll10k_stage$datasetName[mlPerfAll10k_stage$datasetName == "rep200FungiDecontamV2SpeciesVSNM_Path_PT"] <- "VSNM species decontaminated"
mlPerfAll10k_stage$datasetName <- factor(mlPerfAll10k_stage$datasetName,
levels = c("VSNM species decontaminated"))
#-------------------------Plot performance-------------------------#
mlPerfAll10k_stage %>%
distinct() %>% droplevels() %>%
reshape2::melt(id.vars = c("rep","abbrev","diseaseType","sampleType","datasetName","metadataName","minorityClassSize","majorityClassSize","minorityClassName","majorityClassName","nullAUPR","nullAUROC")) %>%
summarySE(measurevar = "value", groupvars = c("datasetName","metadataName","variable","abbrev","minorityClassSize","majorityClassSize","minorityClassName","majorityClassName","nullAUPR","nullAUROC")) %>%
mutate(nullAUPR = ifelse(variable=="AUROC",NA,nullAUPR), nullAUROC = ifelse(variable=="AUPR",NA,nullAUROC)) %>%
ggplot(aes(reorder(abbrev, value, FUN=median),value, color=variable)) +
geom_errorbar(aes(ymin=ifelse(value-ci<0,0,value-ci), ymax=ifelse(value+ci>1,1,value+ci)),width=0.4,size=0.6,position = position_dodge(0.9)) +
geom_errorbar(aes(y=nullAUPR,ymin=nullAUPR,ymax=nullAUPR),lty="dotted",position = position_dodge(0.9)) +
geom_errorbar(aes(y=nullAUROC,ymin=nullAUROC,ymax=nullAUROC),lty="dotted",position = position_dodge(0.9)) +
geom_point(position = position_dodge(0.9), size=1.5) + xlab("Cancer type") + ylab("Area Under Curve") + theme_pubr() +
scale_y_continuous(breaks = seq(0, 1, by = 0.1), limits = c(0,1)) +
ggtitle("Stage I vs Stage IV | Intratumoral | Species") + theme(plot.title = element_text(hjust = 0.5)) +
rotate_x_text(0) + scale_color_nejm(name = "Features") + geom_hline(yintercept = 1, linetype="dashed")
ggsave("Figures/Supplementary_Figures/mlPerfAll10k_rep1_tcga_stage_decontamV2.pdf", dpi = "retina",
width = 6, height = 4, units = "in")
#----------------------------------------------------------#
# Plot results for blood derived normals - VSNM data
#----------------------------------------------------------#
source("Supporting_scripts/S00-SummarySE.R") # Contains a function that calculates std error and 95% confidence intervals
mlPerfAll10k_BDN_EarlyStage_VSNM <- read.csv("Interim_data/rep_perfFungi_10k_rep1_tcga_vsnm_bdn_early_stage_ALL_5Apr22.csv", stringsAsFactors = FALSE)
abbreviationsTCGA_Allcancer <- read.csv("Supporting_data/tcga_abbreviations.csv", stringsAsFactors = FALSE, row.names = 1)
mlPerfAll10k_BDN_EarlyStage_VSNM$abbrev <- abbreviationsTCGA_Allcancer[mlPerfAll10k_BDN_EarlyStage_VSNM$diseaseType,"abbrev"]
mlPerfAll10k_BDN_EarlyStage_VSNM <- mlPerfAll10k_BDN_EarlyStage_VSNM[,!(colnames(mlPerfAll10k_BDN_EarlyStage_VSNM) == "X")]
colnames(mlPerfAll10k_BDN_EarlyStage_VSNM)[1:2] <- c("AUROC","AUPR")
# Add null perf values. Note: AUPR null is prevalence of **positive class**
# For stage, "StageIV" is the positive class and is used to calculate null AUPR
mlPerfAll10k_BDN_EarlyStage_VSNM$nullAUPR <- ifelse(mlPerfAll10k_BDN_EarlyStage_VSNM$minorityClassName == "SolidTissueNormal",
yes=mlPerfAll10k_BDN_EarlyStage_VSNM$majorityClassSize/(mlPerfAll10k_BDN_EarlyStage_VSNM$minorityClassSize+mlPerfAll10k_BDN_EarlyStage_VSNM$majorityClassSize),
no=mlPerfAll10k_BDN_EarlyStage_VSNM$minorityClassSize/(mlPerfAll10k_BDN_EarlyStage_VSNM$minorityClassSize+mlPerfAll10k_BDN_EarlyStage_VSNM$majorityClassSize))
mlPerfAll10k_BDN_EarlyStage_VSNM$nullAUROC <- 0.5
# Rename entries in the "datasetName" column
table(mlPerfAll10k_BDN_EarlyStage_VSNM$datasetName)
mlPerfAll10k_BDN_EarlyStage_VSNM$datasetName[mlPerfAll10k_BDN_EarlyStage_VSNM$datasetName == "rep200FungiDecontamV2SpeciesVSNM_Path_BDN"] <- "VSNM species decontaminated"
mlPerfAll10k_BDN_EarlyStage_VSNM$datasetName <- factor(mlPerfAll10k_BDN_EarlyStage_VSNM$datasetName,
levels = c("VSNM species decontaminated"))
# Species level
mlPerfAll10k_BDN_EarlyStage_VSNM %>%
distinct() %>% droplevels() %>%
reshape2::melt(id.vars = c("rep","abbrev","diseaseType","sampleType","datasetName","metadataName","minorityClassSize","majorityClassSize","minorityClassName","majorityClassName","nullAUPR","nullAUROC")) %>%
summarySE(measurevar = "value", groupvars = c("datasetName","variable","abbrev","metadataName","minorityClassSize","majorityClassSize","minorityClassName","majorityClassName","nullAUPR","nullAUROC","minorityClassSize","majorityClassSize")) %>%
mutate(nullAUPR = ifelse(variable=="AUROC",NA,nullAUPR), nullAUROC = ifelse(variable=="AUPR",NA,nullAUROC)) %>%
ggplot(aes(reorder(abbrev, value, FUN=median),value, color=variable)) +
geom_errorbar(aes(ymin=ifelse(value-ci<0,0,value-ci), ymax=ifelse(value+ci>1,1,value+ci)),width=0.4,size=0.6,position = position_dodge(0.9)) +
geom_errorbar(aes(y=nullAUPR,ymin=nullAUPR,ymax=nullAUPR),lty="dotted",position = position_dodge(0.9)) +
geom_errorbar(aes(y=nullAUROC,ymin=nullAUROC,ymax=nullAUROC),lty="dotted",position = position_dodge(0.9)) +
geom_point(position = position_dodge(0.9), size=1.5) + xlab("Cancer type") + ylab("Area Under Curve") + theme_pubr() +
# facet_wrap(~variable) +
scale_y_continuous(breaks = seq(0, 1, by = 0.1), limits = c(0,1)) +
ggtitle("TCGA Stage Ia-IIc only | VSNM Data | Blood Derived Normal | 1 Vs All | Fungi") + theme(plot.title = element_text(hjust = 0.5)) +
rotate_x_text(90) + scale_color_nejm(name = "Features") + geom_hline(yintercept = 1, linetype="dashed")
ggsave("Figures/Supplementary_Figures/mlPerfAll10k_rep1_VSNM_BDN_early_stage_fungi_decontamV2.svg", dpi = "retina",
width = 8, height = 4, units = "in")
require(gmodels)
mlPerfAll10k_BDN_EarlyStage_VSNM %>%
distinct() %>% droplevels() %>%
pull(AUROC) %>% ci()
# Estimate CI lower CI upper Std. Error
# 0.912230419 0.895167944 0.929292893 0.008616977
mlPerfAll10k_BDN_EarlyStage_VSNM %>%
distinct() %>% droplevels() %>%
filter(abbrev == "BRCA") %>%
pull(AUROC) %>% ci()
# Estimate CI lower CI upper Std. Error
# 0.995297271 0.992019708 0.998574834 0.001448866
#----------------------------------------------------------#
# Plot results for blood derived normals - Raw data
#----------------------------------------------------------#
source("Supporting_scripts/S00-SummarySE.R") # Contains a function that calculates std error and 95% confidence intervals
mlPerfAll10k_BDN_EarlyStage_Raw <- read.csv("Interim_data/rep_perfFungi_10k_rep1_tcga_bdn_early_stage_by_seq_center_ALL_DecontamV2_5Apr22.csv", stringsAsFactors = FALSE)
abbreviationsTCGA_Allcancer <- read.csv("Supporting_data/tcga_abbreviations.csv", stringsAsFactors = FALSE, row.names = 1)
mlPerfAll10k_BDN_EarlyStage_Raw$abbrev <- abbreviationsTCGA_Allcancer[mlPerfAll10k_BDN_EarlyStage_Raw$diseaseType,"abbrev"]
mlPerfAll10k_BDN_EarlyStage_Raw <- mlPerfAll10k_BDN_EarlyStage_Raw[,!(colnames(mlPerfAll10k_BDN_EarlyStage_Raw) == "X")]
colnames(mlPerfAll10k_BDN_EarlyStage_Raw)[1:2] <- c("AUROC","AUPR")
# Add null perf values. Note: AUPR null is prevalence of **positive class**
# For stage, "StageIV" is the positive class and is used to calculate null AUPR
mlPerfAll10k_BDN_EarlyStage_Raw$nullAUPR <- ifelse(mlPerfAll10k_BDN_EarlyStage_Raw$minorityClassName == "SolidTissueNormal",
yes=mlPerfAll10k_BDN_EarlyStage_Raw$majorityClassSize/(mlPerfAll10k_BDN_EarlyStage_Raw$minorityClassSize+mlPerfAll10k_BDN_EarlyStage_Raw$majorityClassSize),
no=mlPerfAll10k_BDN_EarlyStage_Raw$minorityClassSize/(mlPerfAll10k_BDN_EarlyStage_Raw$minorityClassSize+mlPerfAll10k_BDN_EarlyStage_Raw$majorityClassSize))
mlPerfAll10k_BDN_EarlyStage_Raw$nullAUROC <- 0.5
# Rename entries in the "datasetName" column
table(mlPerfAll10k_BDN_EarlyStage_Raw$datasetName)
mlPerfAll10k_BDN_EarlyStage_Raw$datasetName[mlPerfAll10k_BDN_EarlyStage_Raw$datasetName == "rep200_HiSeq_Fungi_DecontamV2_Path_BDN_HMS"] <- "HMS species decontaminated (WGS)"
mlPerfAll10k_BDN_EarlyStage_Raw$datasetName[mlPerfAll10k_BDN_EarlyStage_Raw$datasetName == "rep200_HiSeq_Fungi_DecontamV2_Path_BDN_BCM"] <- "BCM species decontaminated (WGS)"
mlPerfAll10k_BDN_EarlyStage_Raw$datasetName[mlPerfAll10k_BDN_EarlyStage_Raw$datasetName == "rep200_HiSeq_Fungi_DecontamV2_Path_BDN_MDA"] <- "MDA species decontaminated (WGS)"
mlPerfAll10k_BDN_EarlyStage_Raw$datasetName[mlPerfAll10k_BDN_EarlyStage_Raw$datasetName == "rep200_HiSeq_Fungi_DecontamV2_Path_BDN_Broad_WGS"] <- "Broad species decontaminated (WGS)"
mlPerfAll10k_BDN_EarlyStage_Raw$datasetName <- factor(mlPerfAll10k_BDN_EarlyStage_Raw$datasetName,
levels = c("HMS species decontaminated (WGS)",
"BCM species decontaminated (WGS)",
"MDA species decontaminated (WGS)",
"Broad species decontaminated (WGS)"))
# Species level
mlPerfAll10k_BDN_EarlyStage_Raw %>%
# filter(grepl("MDA",datasetName)) %>%
distinct() %>% droplevels() %>%
reshape2::melt(id.vars = c("rep","abbrev","diseaseType","sampleType","datasetName","metadataName","minorityClassSize","majorityClassSize","minorityClassName","majorityClassName","nullAUPR","nullAUROC")) %>%
summarySE(measurevar = "value", groupvars = c("datasetName","variable","abbrev","metadataName","minorityClassSize","majorityClassSize","minorityClassName","majorityClassName","nullAUPR","nullAUROC","minorityClassSize","majorityClassSize")) %>%
mutate(nullAUPR = ifelse(variable=="AUROC",NA,nullAUPR), nullAUROC = ifelse(variable=="AUPR",NA,nullAUROC)) %>%
ggplot(aes(reorder(abbrev, value, FUN=median),value, color=datasetName)) +
geom_errorbar(aes(ymin=ifelse(value-ci<0,0,value-ci), ymax=ifelse(value+ci>1,1,value+ci)),width=0.4,size=0.6,position = position_dodge(0.9)) +
geom_errorbar(aes(y=nullAUPR,ymin=nullAUPR,ymax=nullAUPR),lty="solid",position = position_dodge(0.9)) +
geom_errorbar(aes(y=nullAUROC,ymin=nullAUROC,ymax=nullAUROC),lty="dotted",position = position_dodge(0.9)) +
geom_point(position = position_dodge(0.9), size=1.5) + xlab("Cancer type") + ylab("Area Under Curve") + theme_pubr() +
facet_wrap(~variable) +
guides(color=guide_legend(nrow=2, byrow=TRUE)) +
scale_y_continuous(breaks = seq(0, 1, by = 0.1), limits = c(0,1)) +
ggtitle("TCGA Stage Ia-IIc only | Raw Data | Blood Derived Normal | 1 Vs All | Fungi") + theme(plot.title = element_text(hjust = 0.5)) +
rotate_x_text(90) + scale_color_nejm(name = "Features") + geom_hline(yintercept = 1, linetype="dashed")
ggsave("Figures/Supplementary_Figures/mlPerfAll10k_rep1_Raw_BDN_early_stage_fungi_decontamV2.svg", dpi = "retina",
width = 8, height = 4, units = "in")
require(gmodels)
mlPerfAll10k_BDN_EarlyStage_Raw %>%
distinct() %>% droplevels() %>%
pull(AUROC) %>% ci()
# Estimate CI lower CI upper Std. Error
# 0.88372927 0.86255872 0.90489981 0.01069166