Clustering_Time_Series-Bivariant-CCF
Gonzalo Aris 16021
2023-03-05
Libraries
library(tidyr)
library(readr)
library(ggplot2) # ggplot graphs
library(knitr)
library(readxl)
library(xlsx)
library(openxlsx)
library(reactable) # reactable(df)
library(naniar) # miss_case_summary
library(dplyr)
## KNN imputation
library(caret)
library(RANN)
# CLustering
library(factoextra) # Clustering visualization
library(cluster) # Clustering algorithms
library(dendextend) # For comparing two dendrograms
library(corrplot) # Corelation between dendrograms
library(tidyverse) # Data manupulation
library(NbClust) # Determine optimal no. of clusters [not working...]
library(TSclust)
library(mclust) # Adjusted Rand index
#RandomForest
library(randomForest) # RandomForest Discrete Classification
library(imbalance) # To create a more balanced datasetFunctions
source("../../scripts/useful-functions/get_column_position.R")
# In a normal script it will be: source("./scripts/useful-functions/get_column_position.R")Reading Data
Time Series Data: Heart Rate and SatO2
cuantiles_TS_HR_P2 = data.frame(read_xlsx("../../data/clean-data/BoxBasedImputation/cuantiles_TS_HR_valid_patients_input_P2.xlsx", sheet = "FC_valid_patients_input_P2" ))
SatO2_TS_HR_P2 = data.frame(read_xlsx("../../data/clean-data/BoxBasedImputation/SatO2_valid_patients_input_P2.xlsx", sheet = "SatO2_valid_patients_input_P2" ))
FC_TS_HR_P2 = data.frame(read_xlsx("../../data/clean-data/BoxBasedImputation/FC_valid_patients_input_P2.xlsx", sheet = "FC_valid_patients_input_P2" ))
# First patients with OAF
name_patients_DETERIORO_OAF_0 <- data.frame(read_csv("../../data/clean-data/name_patients_DETERIORO_OAF_0.csv"))
name_patients_DETERIORO_OAF_0 <- name_patients_DETERIORO_OAF_0$x
name_patients_DETERIORO_OAF_0_8 <- data.frame(read_csv("../../data/clean-data/name_patients_DETERIORO_OAF_0_8.csv"))
name_patients_DETERIORO_OAF_0_8 <- name_patients_DETERIORO_OAF_0_8$x
## Deterioro and Not deterioro
file_patient_name_NO_DETERIORO <- data.frame(read_csv("../../data/info-patients/file_patient_name_NO_DETERIORO.csv"))
file_patient_name_NO_DETERIORO <- file_patient_name_NO_DETERIORO$x
file_patient_name_DETERIORO <- data.frame(read_csv("../../data/info-patients/file_patient_name_DETERIORO.csv"))
file_patient_name_DETERIORO <- file_patient_name_DETERIORO$xDescriptive Data for Discriminant analysis
valid_patients_P2 <- data.frame(read_xlsx("../../data/clean-data/valid_patients_P2.xlsx"))
valid_patients_P2 <- valid_patients_P2$x
file_patient_name <- data.frame(read_csv("../../data/clean-data/file_patient_name.csv", show_col_types = FALSE))
file_patient_name <- file_patient_name$x
df1 <- data.frame(read_xlsx("../../data/clean-data/descriptive-data/descriptive_data.xlsx"))
rownames(df1) <- file_patient_name
df1 <- df1[valid_patients_P2,]## Delete first patients
valid_patients_P2 <- valid_patients_P2[! valid_patients_P2 %in% union(name_patients_DETERIORO_OAF_0,name_patients_DETERIORO_OAF_0_8)]
cuantiles_TS_HR_P2 = cuantiles_TS_HR_P2[,valid_patients_P2]
SatO2_TS_HR_P2 = cuantiles_TS_HR_P2[,valid_patients_P2]
FC_TS_HR_P2 = cuantiles_TS_HR_P2[,valid_patients_P2]Descriptive Data
df_descriptive <- data.frame(read_xlsx("../../data/clean-data/descriptive-data/descriptive_data_imputed.xlsx"), row.names = TRUE)
rownames(df_descriptive) <- file_patient_namedf_descriptive <- df_descriptive %>% select(-c(FR_8_16h, FR_16_24h, FLUJO2_8_16h,FLUJO2_16_24h,SCORE_WOOD_DOWNES_24H,SAPI_16_24h, SAPI_8_16h))# Class
pos_1 = get_column_position(df_descriptive,"SAPI_0_8h")
pos_2 = get_column_position(df_descriptive,"PAUSAS_APNEA")
df_descriptive[,c(pos_1:pos_2)] <- lapply(df_descriptive[,c(pos_1:pos_2)], as.factor)
#lapply(df_descriptive,class)
df_descriptive <- df_descriptive[valid_patients_P2,]Create a dataframe with CCF [Heart Rate and SatO2]
dimension_col <- dim(FC_TS_HR_P2)[2]
dimension_row <- 480 #lag.max -1SatO2_FC_CCF <- data.frame(matrix(nrow = dimension_row * 2 - 1, ncol = dimension_col))
colnames(SatO2_FC_CCF) <- names(FC_TS_HR_P2)[1:dimension_col]m <- forecast::Ccf(FC_TS_HR_P2[[1]], SatO2_TS_HR_P2[[1]], lag.max = dimension_row - 1, plot = FALSE, drop.lag.0 = FALSE, type = "correlation", ylab = "CCF")for (i in names(SatO2_FC_CCF)) {
ccf_result <- forecast::Ccf(FC_TS_HR_P2[[i]], SatO2_TS_HR_P2[[i]], lag.max = dimension_row - 1, plot = FALSE, drop.lag.0 = FALSE, type = "correlation", ylab = "CCF")
SatO2_FC_CCF[, i] <- ccf_result$acf
}CCFidean Distance first 100 CCF
distance <- dist(t(SatO2_FC_CCF[c(1:101),]), method = "euclidean")
distance_matrix_CCF <- as.matrix(distance)
DD_CCF <- distanceAgnes study
To find which hierarchical clustering methods that can identify stronger clustering structures. Here we see that Ward’s method identifies the strongest clustering structure of the four methods assessed.
datos = t(SatO2_FC_CCF[c(1:101),])
datos_CCF = data.frame(t(SatO2_FC_CCF[c(1:101),]))
#method to assess
m <- c("average", "single","complete","ward")
names(m) <- c("average", "single","complete","ward.D2")
#function to compute coefficient
ac <- function(x){agnes(datos, method = x)$ac}
map_dbl(m,ac)## average single complete ward.D2
## 0.7992972 0.7091074 0.8910785 0.9437709NbClust study
This package will help us identify the optimum number of clusters
based our criteria in the silhouette index
diss_matrix<- DD_CCF
res<-NbClust(datos_CCF, diss=diss_matrix, distance = NULL, min.nc=2, max.nc=5, method = "ward.D2", index = "silhouette")
res$All.index## 2 3 4 5
## 0.5024 0.3152 0.2589 0.2670res$Best.nc## Number_clusters Value_Index
## 2.0000 0.5024#res$Best.partitionhcintper_CCF <- hclust(DD_CCF , "ward.D2")
fviz_dend(hcintper_CCF, palette = "jco",
rect = TRUE, show_labels = FALSE, k = 2)## Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
## of ggplot2 3.3.4.
## ℹ The deprecated feature was likely used in the factoextra package.
## Please report the issue at <https://github.com/kassambara/factoextra/issues>.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
DDclust_CCF <- cutree( hclust(DD_CCF , "ward.D2"), k = 2)fviz_cluster(list(data = datos_CCF, cluster = DDclust_CCF ))
fviz_silhouette(silhouette(DDclust_CCF , DD_CCF ))## cluster size ave.sil.width
## 1 1 49 0.48
## 2 2 9 0.61
Contingency CCF lag.max = 100
DETERIORO_CLUST <- union(intersect(file_patient_name_DETERIORO,names_1),intersect(file_patient_name_DETERIORO,names_2)) # DET and names_1 + DET and names_2, not all the patients are selected so DETERIORO_CLUST represents the DETERIORO patients that we are working with.
NO_DETERIORO_CLUST <- union(intersect(file_patient_name_NO_DETERIORO,names_1),intersect(file_patient_name_NO_DETERIORO,names_2)) #NO_DET and names_1 + NO_DET and names_2, not all the patients are selected so NO_DETERIORO_CLUST represents the NO_DETERIORO patients that we are working with.#DETERIORO
DETERIORO_patients = data.frame(t(rep("#4A235A", length(DETERIORO_CLUST))))
colnames(DETERIORO_patients)<- DETERIORO_CLUST # Assign a color to all DETERIORO patients
#NO DETERIORO
NO_DETERIORO_patients = data.frame(t(rep("#117864", length(NO_DETERIORO_CLUST))))
colnames(NO_DETERIORO_patients)<- NO_DETERIORO_CLUST # Assign a color to all NO_DETERIORO patients
COLOR_CCF <- cbind(DETERIORO_patients,NO_DETERIORO_patients)fviz_dend(hcintper_CCF, k = 2,
k_colors = c("blue", "green3"),
label_cols = as.vector(COLOR_CCF[names(DD_CCF)[hcintper_CCF$order]]), cex = 0.6) 
n1 = length(intersect(file_patient_name_DETERIORO,names_1))
n2 = length(intersect(file_patient_name_DETERIORO,names_2))
n3 = length(intersect(file_patient_name_NO_DETERIORO,names_1))
n4 = length(intersect(file_patient_name_NO_DETERIORO,names_2))
conttingency_table <- data.frame("CLust1" = c(n1,n3), "Clust2" = c(n2,n4))
rownames(conttingency_table) <- c("DETERIORO","NO DETERIORO")
knitr::kable(conttingency_table, align = "lccrr")| CLust1 | Clust2 | |
|---|---|---|
| DETERIORO | 5 | 1 |
| NO DETERIORO | 44 | 8 |
conttingency_table_prop <- data.frame(c(n1,n3)/(n1+n3),c(n2,n4)/(n2+n4))
rownames(conttingency_table_prop) <- c("DETERIORO","NO DETERIORO")
colnames(conttingency_table_prop) <- c("Clust1","Clust2")
knitr::kable(conttingency_table_prop, align = "lccrr")| Clust1 | Clust2 | |
|---|---|---|
| DETERIORO | 0.1020408 | 0.1111111 |
| NO DETERIORO | 0.8979592 | 0.8888889 |
Random Forest: Discriminant TSCLust CCF
data_frame1_CCF = data.frame("CLUSTER" = DDclust_CCF)
data_frame2 = df_descriptive
data_frame_merge_CCF <-
merge(data_frame1_CCF, data_frame2, by = 'row.names', all = TRUE)
data_frame_merge_CCF <- data_frame_merge_CCF[, 2:dim(data_frame_merge_CCF)[2]]
data_frame_merge_CCF$CLUSTER = factor(data_frame_merge_CCF$CLUSTER)table(data_frame_merge_CCF$CLUSTER)##
## 1 2
## 49 9data_frame_merge_CCF[,c(1:dim(data_frame_merge_CCF)[2])]<- lapply(data_frame_merge_CCF[,c(1:dim(data_frame_merge_CCF)[2])], as.numeric)
head(data_frame_merge_CCF)## CLUSTER EDAD PESO EG FR_0_8h FLUJO2_0_8H DIAS_GN DIAS_O2_TOTAL DIAS_OAF
## 1 1 10.0 8.20 41 48 2.00 3 3 0
## 2 1 13.0 7.78 40 56 2.00 2 2 0
## 3 1 3.1 5.66 37 44 1.00 4 4 0
## 4 2 5.3 8.44 38 65 0.40 3 3 0
## 5 1 15.0 7.00 34 37 2.00 4 4 0
## 6 1 1.6 3.80 37 42 0.94 4 4 0
## SAPI_0_8h SCORE_CRUCES_INGRESO SCORE_WOOD_DOWNES_INGRESO SEXO PALIVIZUMAB LM
## 1 3 3 6 1 1 2
## 2 4 4 8 1 1 1
## 3 3 3 7 1 1 2
## 4 4 3 6 1 1 2
## 5 1 3 6 1 2 1
## 6 2 4 7 1 1 2
## DERMATITIS ALERGIAS TABACO ENFERMEDAD_BASE RADIOGRAFIA ANALITICA SUERO
## 1 1 2 1 1 1 1 1
## 2 1 2 2 2 1 1 2
## 3 1 1 1 1 1 1 1
## 4 1 1 1 1 1 1 1
## 5 1 1 2 2 1 1 2
## 6 1 1 2 2 1 1 1
## ETIOLOGIA PREMATURIDAD ALIMENTACION SNG GN_INGRESO OAF OAF_AL_INGRESO
## 1 2 1 2 1 2 1 1
## 2 1 1 1 1 2 1 1
## 3 2 1 2 1 2 1 1
## 4 2 1 2 1 1 1 1
## 5 2 2 2 1 2 1 1
## 6 1 1 2 1 1 1 1
## OAF_TRAS_INGRESO UCIP DETERIORO PAUSAS_APNEA
## 1 1 1 1 1
## 2 1 1 1 1
## 3 1 1 1 1
## 4 1 1 1 1
## 5 1 1 1 1
## 6 1 1 1 1data_frame_merge_CCF$CLUSTER <- factor(data_frame_merge_CCF$CLUSTER)
newSMOTE_CCF <- oversample(data_frame_merge_CCF, ratio = 0.85, method = "SMOTE", classAttr = "CLUSTER")
newSMOTE_CCF <- data.frame(newSMOTE_CCF)
pos_1 <- get_column_position(newSMOTE_CCF, "SAPI_0_8h")
pos_2 <- get_column_position(newSMOTE_CCF, "PAUSAS_APNEA")
columns_to_round <- c(pos_1:pos_2)
newSMOTE_CCF[, columns_to_round] <- lapply(newSMOTE_CCF[, columns_to_round], function(x) round(x, 1))
table(newSMOTE_CCF$CLUSTER)##
## 1 2
## 49 42set.seed(123)
pos_1 = get_column_position(newSMOTE_CCF, "SAPI_0_8h")
pos_2 = get_column_position(newSMOTE_CCF, "PAUSAS_APNEA")
col_names_factor <- names(newSMOTE_CCF[pos_1:pos_2])
newSMOTE_CCF[col_names_factor] <- lapply(newSMOTE_CCF[col_names_factor] , factor)
RF_CCF <- randomForest(CLUSTER ~ ., data = newSMOTE_CCF)
print(RF_CCF)##
## Call:
## randomForest(formula = CLUSTER ~ ., data = newSMOTE_CCF)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 5
##
## OOB estimate of error rate: 10.99%
## Confusion matrix:
## 1 2 class.error
## 1 44 5 0.1020408
## 2 5 37 0.1190476Importance
kable(RF_CCF$importance[order(RF_CCF$importance, decreasing = TRUE),])| x | |
|---|---|
| SCORE_WOOD_DOWNES_INGRESO | 9.3362067 |
| SCORE_CRUCES_INGRESO | 6.5509986 |
| SAPI_0_8h | 4.3842712 |
| ETIOLOGIA | 2.8116091 |
| PESO | 2.3407629 |
| RADIOGRAFIA | 2.2294075 |
| SEXO | 1.6634895 |
| DIAS_GN | 1.4541484 |
| DIAS_O2_TOTAL | 1.4505291 |
| EDAD | 1.4445559 |
| LM | 1.4356390 |
| FR_0_8h | 1.2697476 |
| GN_INGRESO | 1.2292744 |
| TABACO | 1.2020252 |
| FLUJO2_0_8H | 1.1696037 |
| EG | 0.9153303 |
| ALIMENTACION | 0.7486863 |
| PREMATURIDAD | 0.6391867 |
| ANALITICA | 0.4853217 |
| SNG | 0.3521633 |
| ENFERMEDAD_BASE | 0.2711781 |
| DIAS_OAF | 0.2377083 |
| SUERO | 0.2108443 |
| ALERGIAS | 0.1217301 |
| UCIP | 0.1028728 |
| DETERIORO | 0.0966294 |
| PALIVIZUMAB | 0.0934120 |
| OAF_TRAS_INGRESO | 0.0860743 |
| PAUSAS_APNEA | 0.0802490 |
| OAF | 0.0603296 |
| DERMATITIS | 0.0488441 |
| OAF_AL_INGRESO | 0.0000000 |
Importance of first 50 CCF
data_frame1_CCF = data.frame("CLUSTER" = DDclust_CCF)
data_frame2 = data.frame(t(SatO2_FC_CCF[c(1:101),]))
data_frame_merge_CCF <-
merge(data_frame1_CCF, data_frame2, by = 'row.names', all = TRUE)
data_frame_merge_CCF <- data_frame_merge_CCF[, 2:dim(data_frame_merge_CCF)[2]]set.seed(123)
data_frame_merge_CCF$CLUSTER <- as.factor(data_frame_merge_CCF$CLUSTER)
RF_0 <- randomForest(CLUSTER ~ ., data = data_frame_merge_CCF)
print(RF_0)##
## Call:
## randomForest(formula = CLUSTER ~ ., data = data_frame_merge_CCF)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 10
##
## OOB estimate of error rate: 0%
## Confusion matrix:
## 1 2 class.error
## 1 49 0 0
## 2 0 9 0plot(RF_0$importance, type = "h")
CCF by clusters
plot_data_CCF <- data.frame(t(SatO2_FC_CCF[c(1:101),]))
cluster_data_CCF <- data.frame(DDclust_CCF)
plotting_CCF <- cbind(plot_data_CCF, cluster_data_CCF)
head(plotting_CCF)## X1 X2 X3 X4
## ACR_11231843 0.0004335791 0.0026800010 0.0039622812 0.0046441295
## ADAO_11159808 0.0001094381 -0.0003372668 0.0003987779 0.0002092462
## AGG_11236448 -0.0013688104 -0.0025258245 -0.0032101939 -0.0042764317
## AHL_11239959 -0.0023989337 -0.0052266590 -0.0054464697 -0.0081471913
## AJGD_11119689 -0.0003576793 -0.0008162598 -0.0005385093 -0.0007230021
## AMP_11228639 0.0002634324 0.0006663009 -0.0005199333 0.0008299215
## X5 X6 X7 X8
## ACR_11231843 7.203981e-03 0.007061645 0.0076842907 7.630654e-03
## ADAO_11159808 -1.623079e-04 0.001405437 0.0020338192 -2.433889e-03
## AGG_11236448 -4.098981e-03 -0.003913110 -0.0044643766 -3.812243e-03
## AHL_11239959 -1.182286e-02 -0.015537484 -0.0182401880 -2.248151e-02
## AJGD_11119689 2.780979e-03 0.002467757 0.0009709092 -5.704056e-05
## AMP_11228639 7.965372e-05 0.002581463 0.0009556513 -5.763339e-04
## X9 X10 X11 X12
## ACR_11231843 0.0011541477 -0.0014977354 0.0015747985 0.0027412959
## ADAO_11159808 0.0034024126 0.0005693701 0.0004382390 0.0028263403
## AGG_11236448 -0.0034884464 -0.0018366516 -0.0030502209 -0.0030932730
## AHL_11239959 -0.0246426111 -0.0257563218 -0.0303770734 -0.0305383603
## AJGD_11119689 0.0012180638 0.0008889351 0.0002642562 -0.0005845125
## AMP_11228639 0.0006360153 -0.0021070910 -0.0001457457 -0.0006286650
## X13 X14 X15 X16
## ACR_11231843 -0.0012563791 -0.0072490300 -0.0061775054 -0.0042857758
## ADAO_11159808 0.0013409788 0.0019777046 0.0012957525 0.0004308224
## AGG_11236448 0.0024329065 0.0022840021 0.0006671297 0.0024455144
## AHL_11239959 -0.0300882314 -0.0399208646 -0.0364802586 -0.0390877414
## AJGD_11119689 -0.0010046334 -0.0003411414 0.0012670173 -0.0019547537
## AMP_11228639 -0.0005184107 0.0011136237 0.0028704016 0.0011150101
## X17 X18 X19 X20
## ACR_11231843 -3.636382e-03 -0.003516933 -0.0027122897 -0.001284843
## ADAO_11159808 -3.576277e-03 0.001442159 0.0006848414 -0.000669266
## AGG_11236448 -4.504692e-05 0.003706636 0.0041543671 0.003884811
## AHL_11239959 -4.355986e-02 -0.040192137 -0.0427527366 -0.039106720
## AJGD_11119689 2.037051e-03 -0.001724106 -0.0010821281 -0.005371884
## AMP_11228639 4.760198e-03 -0.001430710 0.0035563337 -0.001026648
## X21 X22 X23 X24
## ACR_11231843 -0.0019719796 -0.003068912 -0.0007996714 0.001418854
## ADAO_11159808 -0.0008796252 0.003904162 -0.0037915855 0.003203702
## AGG_11236448 0.0048978034 0.002993458 0.0077458111 0.008945708
## AHL_11239959 -0.0367113603 -0.033335049 -0.0363302775 -0.031059899
## AJGD_11119689 -0.0021673785 0.006887099 -0.0070032804 -0.009363823
## AMP_11228639 -0.0014857750 0.004153272 -0.0021751878 0.003863868
## X25 X26 X27 X28 X29
## ACR_11231843 0.0009437696 0.003782053 0.010887237 0.013280146 0.016968351
## ADAO_11159808 -0.0007678796 -0.001109457 0.003944547 0.005774560 -0.009068616
## AGG_11236448 0.0057222462 0.017769404 0.028777302 0.038949199 0.035914986
## AHL_11239959 -0.0352128645 -0.036601861 -0.043029479 -0.048333482 -0.059119791
## AJGD_11119689 -0.0081886603 -0.011067657 -0.008622958 -0.010048992 -0.002165678
## AMP_11228639 0.0003253959 0.008311592 0.006560036 0.002216409 0.009591607
## X30 X31 X32 X33 X34
## ACR_11231843 0.0136167903 0.008385831 0.010921905 0.008648812 0.006103761
## ADAO_11159808 0.0003874439 -0.002466101 -0.009799162 -0.008931975 -0.016326710
## AGG_11236448 0.0362769742 0.040192627 0.035510527 0.032862108 0.037695124
## AHL_11239959 -0.0634132755 -0.074786468 -0.068685563 -0.079150534 -0.077537483
## AJGD_11119689 -0.0049804151 0.008837385 -0.005148626 -0.013302486 0.002273300
## AMP_11228639 -0.0004985741 0.002128356 0.008011105 -0.001738909 0.006397814
## X35 X36 X37 X38 X39
## ACR_11231843 0.0012823704 -0.001064345 0.005539160 0.001633917 0.002720487
## ADAO_11159808 -0.0122746108 -0.022561360 -0.013583560 -0.027417224 -0.018211320
## AGG_11236448 0.0410813463 0.051974296 0.052692149 0.061916401 0.074867491
## AHL_11239959 -0.0821848400 -0.074948409 -0.074274509 -0.077629529 -0.078331075
## AJGD_11119689 -0.0183615004 -0.018595368 0.002216751 0.004792426 0.003452190
## AMP_11228639 0.0006064979 0.007679668 0.009109502 0.015024574 0.025892060
## X40 X41 X42 X43 X44
## ACR_11231843 0.002095688 -0.0007767707 0.007790719 0.008091110 0.01025183
## ADAO_11159808 -0.019288467 -0.0219156396 -0.033166034 -0.026608065 -0.02464164
## AGG_11236448 0.095602206 0.1005956040 0.095352375 0.096222011 0.08995313
## AHL_11239959 -0.083351334 -0.0829452806 -0.088129677 -0.092405923 -0.09412752
## AJGD_11119689 -0.013032020 -0.0104414413 -0.016322045 -0.002454743 -0.01615293
## AMP_11228639 0.014230001 0.0188068271 0.015089930 0.008690895 0.01359409
## X45 X46 X47 X48 X49
## ACR_11231843 0.011461143 0.005265604 -0.006500759 -0.01653930 -0.019145954
## ADAO_11159808 -0.029244691 -0.022192961 -0.016925910 -0.01309521 -0.006670672
## AGG_11236448 0.075513576 0.072721718 0.074447261 0.08080160 0.096337513
## AHL_11239959 -0.095718411 -0.101214713 -0.099351766 -0.09540954 -0.097778015
## AJGD_11119689 -0.002605373 0.017402271 0.021764256 0.04339378 0.026945155
## AMP_11228639 0.017587160 0.022144110 0.025556164 0.02227766 0.026539077
## X50 X51 X52 X53 X54
## ACR_11231843 -0.01328539 0.0466660069 0.078539004 0.003436640 -0.040578171
## ADAO_11159808 -0.01122056 -0.0192040012 -0.008480597 -0.015677095 -0.015684548
## AGG_11236448 0.11275699 0.1202005447 0.121362273 0.118260219 0.113192804
## AHL_11239959 -0.10259836 -0.0975660015 -0.097635034 -0.099399656 -0.104142381
## AJGD_11119689 0.01736604 -0.0008852263 0.003920554 0.001447258 0.005990889
## AMP_11228639 0.01741680 0.0255593119 0.017152138 0.020785147 0.019921068
## X55 X56 X57 X58 X59
## ACR_11231843 -0.03923186 -0.03843734 -0.03185135 -0.03375636 -0.03155462
## ADAO_11159808 -0.02824683 -0.02761742 -0.03231723 -0.03411898 -0.04269926
## AGG_11236448 0.10657797 0.09517184 0.10000647 0.08734770 0.06026680
## AHL_11239959 -0.10927454 -0.10858498 -0.10513668 -0.11227425 -0.11115260
## AJGD_11119689 0.01952905 0.02519347 0.02603430 0.01601270 0.03185822
## AMP_11228639 0.02036761 0.02478283 0.01488865 0.02440095 0.01461187
## X60 X61 X62 X63 X64
## ACR_11231843 -0.01614012 0.017230667 0.019998453 0.0220845508 0.035200236
## ADAO_11159808 -0.03476621 -0.022482786 -0.018676508 -0.0114794245 0.001431606
## AGG_11236448 0.04530848 0.063400777 0.069710921 0.0768386681 0.091449601
## AHL_11239959 -0.11393258 -0.116610055 -0.119626833 -0.1171933414 -0.125775259
## AJGD_11119689 0.02309504 -0.004168741 0.001406862 -0.0003629521 0.015114474
## AMP_11228639 0.01527477 0.021681867 0.033218459 0.0403875606 0.042386444
## X65 X66 X67 X68 X69
## ACR_11231843 -0.014768194 -0.011308070 0.01165559 0.039247108 0.04067433
## ADAO_11159808 0.021322108 0.016238848 0.02015872 0.028304374 0.02867736
## AGG_11236448 0.088309944 0.084025304 0.08346901 0.043776711 0.01928346
## AHL_11239959 -0.125271045 -0.134881079 -0.13846456 -0.138021332 -0.14516406
## AJGD_11119689 -0.007612185 -0.008576289 -0.01310511 0.001155695 0.02072063
## AMP_11228639 0.046761078 0.041478845 0.03415545 0.038253282 0.04345439
## X70 X71 X72 X73 X74
## ACR_11231843 0.054938263 0.05216774 0.02893450 0.04565816 0.016317459
## ADAO_11159808 0.031406913 0.03285366 0.03034807 0.01507817 0.007630274
## AGG_11236448 0.006741361 0.02914898 0.02863798 0.04718177 0.057838065
## AHL_11239959 -0.148491209 -0.15009041 -0.14325691 -0.14896503 -0.148111238
## AJGD_11119689 0.006656645 0.02928379 0.03218864 0.01925793 0.020437607
## AMP_11228639 0.044788032 0.04858658 0.04187590 0.04227944 0.037635781
## X75 X76 X77 X78 X79
## ACR_11231843 -0.003875787 0.0256742079 0.018148296 0.04219174 0.027417819
## ADAO_11159808 0.018474492 0.0009056106 0.010227621 0.01560631 0.009562942
## AGG_11236448 0.071557804 0.0639719008 0.076988919 0.05250665 0.053139109
## AHL_11239959 -0.152727955 -0.1479525477 -0.145243487 -0.14707256 -0.136525385
## AJGD_11119689 0.004447869 0.0176834733 -0.005893729 0.01482346 -0.002029224
## AMP_11228639 0.049409292 0.0438721130 0.046904276 0.04721456 0.034240507
## X80 X81 X82 X83 X84
## ACR_11231843 0.011569835 -2.315659e-02 -0.026266905 0.002209966 0.006781454
## ADAO_11159808 0.005644969 4.665312e-05 0.009562882 0.028231420 0.035291980
## AGG_11236448 0.053179236 6.266205e-02 0.038755605 0.037632958 0.053041903
## AHL_11239959 -0.145777797 -1.373812e-01 -0.134550696 -0.127767682 -0.123160078
## AJGD_11119689 -0.000442201 -8.141466e-03 -0.028236727 -0.021561516 -0.020813623
## AMP_11228639 0.043992079 4.623287e-02 0.049872969 0.055800867 0.051210087
## X85 X86 X87 X88 X89
## ACR_11231843 -0.015863194 -0.0008034249 -0.03327019 -0.030903612 0.01948229
## ADAO_11159808 0.030878793 0.0365097797 0.04058043 0.035934982 0.03357107
## AGG_11236448 0.083892800 0.0716219676 0.08203580 0.081094153 0.07663220
## AHL_11239959 -0.119413282 -0.1126149959 -0.10323170 -0.101827184 -0.10077971
## AJGD_11119689 -0.009651539 -0.0163368826 -0.02784260 0.005658919 -0.01588456
## AMP_11228639 0.057112175 0.0585364164 0.05274634 0.067226000 0.05090611
## X90 X91 X92 X93 X94
## ACR_11231843 -0.008660467 0.006913348 -0.04212957 -0.04023577 -0.041013020
## ADAO_11159808 0.039765226 0.047538257 0.02947733 0.04493890 0.021011893
## AGG_11236448 0.092984113 0.091885797 0.07406543 0.04952749 0.055319778
## AHL_11239959 -0.105786144 -0.102096714 -0.10295521 -0.09423088 -0.108538702
## AJGD_11119689 -0.014811634 -0.023097969 -0.02852232 -0.02476067 -0.007509145
## AMP_11228639 0.067204569 0.058703439 0.07059440 0.07205772 0.065932511
## X95 X96 X97 X98 X99
## ACR_11231843 -0.049501935 -0.04770504 -0.047457036 -0.04491680 -0.044618932
## ADAO_11159808 0.019975653 0.02109524 0.006412441 0.00139112 0.005480962
## AGG_11236448 0.046182840 0.03734793 0.024458350 0.02264219 0.005794925
## AHL_11239959 -0.101857508 -0.10723867 -0.108969619 -0.10277380 -0.102861916
## AJGD_11119689 0.006218254 0.01048821 0.039346488 0.01535082 0.008367036
## AMP_11228639 0.074911235 0.06127015 0.070413020 0.06243588 0.051809416
## X100 X101 DDclust_CCF
## ACR_11231843 -0.045242042 -0.03506496 1
## ADAO_11159808 0.001913931 0.01776448 1
## AGG_11236448 0.027707410 0.01951120 1
## AHL_11239959 -0.111476327 -0.11508806 2
## AJGD_11119689 0.023412887 0.01173931 1
## AMP_11228639 0.056981920 0.05601130 1## Mean by groups
rp_tbl <- aggregate(plotting_CCF, by = list(plotting_CCF$DDclust_CCF), mean)
row.names(rp_tbl) <- paste0("Group",rp_tbl$DDclust_CCF)
rp_tbl <- rp_tbl %>%
select(starts_with('X'))
rp_tbl <- data.frame(t(rp_tbl))
head(rp_tbl)## Group1 Group2
## X1 2.769181e-04 -0.001403052
## X2 5.712380e-04 -0.002722402
## X3 -8.583775e-05 -0.003634944
## X4 4.847546e-04 -0.004069677
## X5 6.010612e-04 -0.006732902
## X6 3.038989e-04 -0.008340906# Create plotting data-frame
CCF_values_by_group <- data.frame("value_CCF" = c(rp_tbl$Group1,rp_tbl$Group2),
"cluster" = c(rep("Group1", times = length(rp_tbl$Group1)),
rep("Group2", times = length(rp_tbl$Group2))),
"index" = c(c(1:length(rp_tbl$Group1)),c(1:length(rp_tbl$Group2))))
p <- ggplot(CCF_values_by_group, aes(x = index, y = value_CCF, group = cluster)) +
geom_line(aes(color=cluster)) +
scale_color_brewer(palette="Paired") + theme_minimal()
p
Adjusted Rand index
cluster_study_CCF <- list(DDclust_CCF)write.csv(cluster_study_CCF, "../../data/clusters/cluster_study_CCF.csv")