Análise Descritiva RNA

# BiBliotecas
library(readxl)
library(astsa)
library(tidyverse)
library(forecast)
library(tidyquant)
library(tscount)
library(gridExtra)

\[ a = 1 \\ c = 2 \\ g = 3 \\ t = 4 \]

1 RNA Australia X China.

res <- Australia_RNA %>% filter(Australia_RNA$RNA_num!=China_Wuhan_RNA$RNA_num) %>% 
  bind_cols(China_Wuhan_RNA  %>% filter(China_Wuhan_RNA$RNA_num!=Australia_RNA$RNA_num)) %>%
  transmute(Tempo = Time, RNA_Australia = RNA_num,
            RNA_China = RNA_num1)

res

M1<-table(Australia_RNA$RNA, China_Wuhan_RNA$RNA)
M1

##    
##        A    C    G    T
##   A 2702 1650 1752 2789
##   C 1576  996 1189 1708
##   G 1719 1032 1058 2038
##   T 2896 1795 1853 3029

test<- chisq.test(x = M1)
test

## 
##  Pearson's Chi-squared test
## 
## data:  M1
## X-squared = 43.399, df = 9, p-value = 1.821e-06

acf(Australia_RNA$RNA_num %>% ts(start = 1, end = nrow(Australia_RNA)), max.lag = 30, main = 'Australia RNA')

pacf(Australia_RNA$RNA_num %>% ts(start = 1, end = nrow(Australia_RNA)), max.lag = 30, main = 'Australia RNA')

ccf(x = Australia_RNA$RNA_num %>% ts(start = 1, end = nrow(Australia_RNA)),
    y = China_Wuhan_RNA$RNA_num %>% ts(start = 1, end = nrow(China_Wuhan_RNA)), lag.max = 30, main = "Autocorrelação Cruzada entre os RNA", type = 'correlation', ylab='CCF(X,Y)')

1.1 Aplicação do modelo

m1<-Logit_ts(y = Australia_RNA$RNA_num, Termo_Ar = c(1), Intercept = T,
             xreg = China_Wuhan_RNA$RNA_num %>% as.matrix())


m1$df_pred %>% filter(fitted != y_real)

m1$accuracy

## [1] 33.78664

m1$pi_matrix_estimate %>% summary()

##        V1               V2               V3               V4        
##  Min.   :0.0000   Min.   :0.0000   Min.   :0.0000   Min.   :0.0000  
##  1st Qu.:0.2530   1st Qu.:0.1491   1st Qu.:0.1839   1st Qu.:0.2635  
##  Median :0.2795   Median :0.1645   Median :0.1931   Median :0.3342  
##  Mean   :0.2986   Mean   :0.1836   Mean   :0.1963   Mean   :0.3214  
##  3rd Qu.:0.3254   3rd Qu.:0.2230   3rd Qu.:0.2604   3rd Qu.:0.3417  
##  Max.   :0.3836   Max.   :0.2302   Max.   :0.2787   Max.   :0.3830

2 RNA Alemanha X China.

res <- Alemanha_RNA %>% filter(Alemanha_RNA$RNA_num!=China_Wuhan_RNA$RNA_num) %>% 
  bind_cols(China_Wuhan_RNA  %>% filter(China_Wuhan_RNA$RNA_num!=Alemanha_RNA$RNA_num)) %>%
  transmute(Tempo = Time, RNA_Alemanha = RNA_num,
            RNA_China = RNA_num1)

res

M1<-table(Alemanha_RNA$RNA, China_Wuhan_RNA$RNA)
M1

##    
##        A    C    G    T
##   A 2673 1622 1726 2875
##   C 1613 1000  983 1877
##   G 1760 1185 1077 1826
##   T 2847 1666 2066 2986

test<- chisq.test(x = M1)
test

## 
##  Pearson's Chi-squared test
## 
## data:  M1
## X-squared = 60.08, df = 9, p-value = 1.294e-09

acf(Alemanha_RNA$RNA_num %>% ts(start = 1, end = nrow(Alemanha_RNA)), max.lag = 30, main = 'Australia RNA')

pacf(Alemanha_RNA$RNA_num %>% ts(start = 1, end = nrow(Alemanha_RNA)), max.lag = 30, main = 'Australia RNA')

ccf(x = Alemanha_RNA$RNA_num %>% ts(start = 1, end = nrow(Alemanha_RNA)),
    y = China_Wuhan_RNA$RNA_num %>% ts(start = 1, end = nrow(China_Wuhan_RNA)), lag.max = 30, main = "Autocorrelação Cruzada entre os RNA", type = 'correlation', ylab='CCF(X,Y)')

2.1 Aplicação do modelo

m2<-Logit_ts(y = Australia_RNA$RNA_num, Termo_Ar = c(1), Intercept = T,
             xreg = China_Wuhan_RNA$RNA_num %>% as.matrix())


m2$df_pred %>% filter(fitted != y_real)

m2$accuracy

## [1] 33.78664

m2$pi_matrix_estimate %>% summary()

##        V1               V2               V3               V4        
##  Min.   :0.0000   Min.   :0.0000   Min.   :0.0000   Min.   :0.0000  
##  1st Qu.:0.2530   1st Qu.:0.1491   1st Qu.:0.1839   1st Qu.:0.2635  
##  Median :0.2795   Median :0.1645   Median :0.1931   Median :0.3342  
##  Mean   :0.2986   Mean   :0.1836   Mean   :0.1963   Mean   :0.3214  
##  3rd Qu.:0.3254   3rd Qu.:0.2230   3rd Qu.:0.2604   3rd Qu.:0.3417  
##  Max.   :0.3836   Max.   :0.2301   Max.   :0.2787   Max.   :0.3830

3 RNA USA X China.

res <- USA_RNA %>% filter(USA_RNA$RNA_num != China_Wuhan_RNA$RNA_num) %>% 
  bind_cols(China_Wuhan_RNA  %>% filter(China_Wuhan_RNA$RNA_num != USA_RNA$RNA_num)) %>%
  transmute(Tempo = Time, RNA_USA = RNA_num, RNA_China = RNA_num1)

res

M1<-table(USA_RNA$RNA, China_Wuhan_RNA$RNA)
M1

##    
##        A    C    G    T
##   A 2602 1682 1769 2839
##   C 1589 1021 1173 1691
##   G 1738  990 1033 2083
##   T 2964 1780 1877 2951

test<- chisq.test(x = M1)
test

## 
##  Pearson's Chi-squared test
## 
## data:  M1
## X-squared = 65.627, df = 9, p-value = 1.089e-10

acf(USA_RNA$RNA_num %>% ts(start = 1, end = nrow(USA_RNA)), max.lag = 30, main = 'USA RNA')

pacf(USA_RNA$RNA_num %>% ts(start = 1, end = nrow(USA_RNA)), max.lag = 30, main = 'USA RNA')

ccf(x = USA_RNA$RNA_num %>% ts(start = 1, end = nrow(USA_RNA)),
    y = China_Wuhan_RNA$RNA_num %>% ts(start = 1, end = nrow(China_Wuhan_RNA)), lag.max = 30, main = "Autocorrelação Cruzada entre os RNA", type = 'correlation', ylab='CCF(X,Y)')

3.1 Aplicação do modelo

m3<-Logit_ts(y = USA_RNA$RNA_num, Termo_Ar = c(1), Intercept = T,
             xreg = China_Wuhan_RNA$RNA_num %>% as.matrix())


m3$df_pred %>% filter(fitted != y_real)

m3$accuracy

## [1] 33.98811

m3$pi_matrix_estimate %>% summary()

##        V1               V2               V3               V4        
##  Min.   :0.0000   Min.   :0.0000   Min.   :0.0000   Min.   :0.0000  
##  1st Qu.:0.2482   1st Qu.:0.1476   1st Qu.:0.1833   1st Qu.:0.2698  
##  Median :0.2765   Median :0.1629   Median :0.1962   Median :0.3313  
##  Mean   :0.2986   Mean   :0.1838   Mean   :0.1962   Mean   :0.3214  
##  3rd Qu.:0.3201   3rd Qu.:0.2266   3rd Qu.:0.2566   3rd Qu.:0.3498  
##  Max.   :0.3849   Max.   :0.2275   Max.   :0.2821   Max.   :0.3911

4 RNA França X China.

res <- Franca_RNA %>% filter(Franca_RNA$RNA_num != China_Wuhan_RNA$RNA_num) %>% 
  bind_cols(China_Wuhan_RNA  %>% filter(China_Wuhan_RNA$RNA_num != Franca_RNA$RNA_num)) %>%
  transmute(Tempo = Time, RNA_Franca = RNA_num, RNA_China = RNA_num1)

res

M1<-table(Franca_RNA$RNA, China_Wuhan_RNA$RNA)
M1

##    
##        A    C    G    T
##   A 2660 1608 1769 2857
##   C 1593 1019 1014 1848
##   G 1795 1148 1025 1878
##   T 2845 1698 2044 2981

test<- chisq.test(x = M1)
test

## 
##  Pearson's Chi-squared test
## 
## data:  M1
## X-squared = 48.989, df = 9, p-value = 1.668e-07

acf(Franca_RNA$RNA_num %>% ts(start = 1, end = nrow(Franca_RNA)), max.lag = 30, main = 'Japão RNA')

pacf(Franca_RNA$RNA_num %>% ts(start = 1, end = nrow(Franca_RNA)), max.lag = 30, main = 'Japão RNA')

ccf(x = Franca_RNA$RNA_num %>% ts(start = 1, end = nrow(Franca_RNA)),
    y = China_Wuhan_RNA$RNA_num %>% ts(start = 1, end = nrow(China_Wuhan_RNA)), lag.max = 30, main = "Autocorrelação Cruzada entre os RNA" , type = 'correlation', ylab='CCF(X,Y)')

4.1 Aplicação do modelo

m4<-Logit_ts(y = Franca_RNA$RNA_num, Termo_Ar = c(1), Intercept = T,
             xreg = China_Wuhan_RNA$RNA_num %>% as.matrix())


m4$df_pred %>% filter(fitted != y_real)

m4$accuracy

## [1] 34.01498

m4$pi_matrix_estimate %>% summary()

##        V1               V2               V3               V4        
##  Min.   :0.0000   Min.   :0.0000   Min.   :0.0000   Min.   :0.0000  
##  1st Qu.:0.2491   1st Qu.:0.1518   1st Qu.:0.1815   1st Qu.:0.2590  
##  Median :0.2753   Median :0.1649   Median :0.1930   Median :0.3358  
##  Mean   :0.2986   Mean   :0.1838   Mean   :0.1963   Mean   :0.3213  
##  3rd Qu.:0.3201   3rd Qu.:0.2212   3rd Qu.:0.2628   3rd Qu.:0.3400  
##  Max.   :0.3804   Max.   :0.2331   Max.   :0.2771   Max.   :0.3804

5 RNA Brasil X China.

res <- Brasil_RNA %>% filter(Brasil_RNA$RNA_num != China_Wuhan_RNA$RNA_num) %>% 
  bind_cols(China_Wuhan_RNA  %>% filter(China_Wuhan_RNA$RNA_num != Brasil_RNA$RNA_num)) %>%
  transmute(Tempo = Time, RNA_Brasil = RNA_num, RNA_China = RNA_num1)

res

M1<-table(Brasil_RNA$RNA, China_Wuhan_RNA$RNA)
M1

##    
##        A    C    G    T
##   A 2740 1601 1741 2810
##   C 1546 1044 1032 1853
##   G 1793 1145 1029 1877
##   T 2814 1683 2050 3024

test<- chisq.test(x = M1)
test

## 
##  Pearson's Chi-squared test
## 
## data:  M1
## X-squared = 55.688, df = 9, p-value = 9.003e-09

acf(Brasil_RNA$RNA_num %>% ts(start = 1, end = nrow(Brasil_RNA)), max.lag = 30, main = 'Brasil RNA')

pacf(Brasil_RNA$RNA_num %>% ts(start = 1, end = nrow(Brasil_RNA)), max.lag = 30, main = 'Brasil RNA')

ccf(x = Brasil_RNA$RNA_num %>% ts(start = 1, end = nrow(Brasil_RNA)),
    y = China_Wuhan_RNA$RNA_num %>% ts(start = 1, end = nrow(China_Wuhan_RNA)), lag.max = 30, main = "Autocorrelação Cruzada entre os RNA" , type = 'correlation', ylab='CCF(X,Y)')

5.1 Aplicação do modelo

m5<-Logit_ts(y = Brasil_RNA$RNA_num, Termo_Ar = c(1), Intercept = T,
             xreg = China_Wuhan_RNA$RNA_num %>% as.matrix())


m5$df_pred %>% filter(fitted != y_real)

m5$accuracy

## [1] 33.98476

m5$pi_matrix_estimate %>% summary()

##        V1               V2               V3               V4        
##  Min.   :0.0000   Min.   :0.0000   Min.   :0.0000   Min.   :0.0000  
##  1st Qu.:0.2520   1st Qu.:0.1534   1st Qu.:0.1812   1st Qu.:0.2618  
##  Median :0.2808   Median :0.1670   Median :0.1932   Median :0.3356  
##  Mean   :0.2986   Mean   :0.1838   Mean   :0.1962   Mean   :0.3213  
##  3rd Qu.:0.3259   3rd Qu.:0.2182   3rd Qu.:0.2628   3rd Qu.:0.3437  
##  Max.   :0.3882   Max.   :0.2360   Max.   :0.2771   Max.   :0.3836