Análisis Marco
2024-08-20
install.packages(c("tidyverse","gtsummary", "readxl", "ggpubr","cowplot"))## Installing packages into '/cloud/lib/x86_64-pc-linux-gnu-library/4.4'
## (as 'lib' is unspecified)# Cargar las librerías necesarias
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(gtsummary)
library(readxl)library(readxl)
data <- read_excel("base de datos polimorfismos gen RAGE.xlsx")
#View(data)# Ensure `data` is a data frame
data <- as.data.frame(data)
# Recodificar las variables de interés incluyendo "Renal"
data <- data %>%
mutate(
Sexo = recode(Sexo, "M" = "Másculino", "F" = "Femenino"))
data$`Peso (kg)` <- as.numeric(data$`Peso (kg)`)summary_table <- data %>%
select(Grup,Sexo, Edad, `Peso (kg)`, `Talla (M)`, `Cintura (cm)`, `Cadera (cm)`,IMC
) %>%
tbl_summary(
by =Grup, # Agrupar por la variable dependiente "Renal_Dysfunction"
missing = "no" # Excluye valores faltantes
) %>% add_overall() %>%
add_p() %>%
modify_header(label = "**Variables**") %>%
modify_caption("**Table: Características Clínicas y Demograficas de pacientes con SD**") %>%
as_gt() %>%
gt::tab_style(
style = gt::cell_text(weight = "bold"),
locations = gt::cells_column_labels()
)
summary_table| Variables | Overall N = 1051 |
CTRL N = 551 |
SD N = 501 |
p-value2 |
|---|---|---|---|---|
| Sexo | 0.6 | |||
| Femenino | 51 (49%) | 28 (51%) | 23 (46%) | |
| Másculino | 54 (51%) | 27 (49%) | 27 (54%) | |
| Edad | 10.0 (7.0, 12.0) | 10.0 (8.0, 12.0) | 10.0 (6.0, 13.0) | 0.8 |
| Peso (kg) | 32 (22, 43) | 31 (24, 41) | 32 (18, 46) | 0.9 |
| Talla (M) | 1.33 (1.17, 1.45) | 1.38 (1.26, 1.48) | 1.28 (1.05, 1.40) | 0.002 |
| Cintura (cm) | 61 (54, 71) | 58 (53, 63) | 65 (56, 75) | 0.005 |
| Cadera (cm) | 72 (62, 84) | 70 (64, 81) | 78 (61, 86) | 0.4 |
| IMC | 17.1 (15.7, 20.5) | 16.5 (15.4, 18.2) | 19.1 (16.8, 24.0) | <0.001 |
| 1 n (%); Median (Q1, Q3) | ||||
| 2 Pearson’s Chi-squared test; Wilcoxon rank sum test | ||||
summary_table <- data %>%
select(Grup, HBA1c, `Acido Urico (mg/dL)`, `Glucosa (mg/dL)`, `Urea (mg/dL)`,
`Creatinina (mg/dL)`,
) %>%
tbl_summary(
by =Grup, # Agrupar por la variable dependiente "Sindrome Down"
missing = "ifany" # Excluye valores faltantes
) %>% add_overall() %>%
add_p() %>%
modify_header(label = "**Variables**") %>%
modify_caption("**Table: Laboratorios química Sanguínea de pacientes con SD**") %>%
as_gt() %>%
gt::tab_style(
style = gt::cell_text(weight = "bold"),
locations = gt::cells_column_labels()
)
summary_table| Variables | Overall N = 1051 |
CTRL N = 551 |
SD N = 501 |
p-value2 |
|---|---|---|---|---|
| HBA1c | 5.10 (4.88, 5.40) | 5.20 (4.90, 5.30) | 5.06 (4.70, 5.50) | 0.6 |
| Acido Urico (mg/dL) | 4.46 (3.70, 5.40) | 3.86 (3.30, 4.50) | 5.35 (4.50, 5.80) | <0.001 |
| Unknown | 4 | 0 | 4 | |
| Glucosa (mg/dL) | 92 (86, 100) | 93 (85, 104) | 92 (86, 99) | 0.3 |
| Urea (mg/dL) | 25 (20, 30) | 23 (17, 29) | 28 (23, 31) | <0.001 |
| Creatinina (mg/dL) | 0.47 (0.39, 0.60) | 0.45 (0.38, 0.53) | 0.50 (0.40, 0.63) | 0.11 |
| 1 Median (Q1, Q3) | ||||
| 2 Wilcoxon rank sum test | ||||
install.packages(c("ggpubr", "cowplot"))## Installing packages into '/cloud/lib/x86_64-pc-linux-gnu-library/4.4'
## (as 'lib' is unspecified)library(ggpubr)
library(cowplot)##
## Attaching package: 'cowplot'## The following object is masked from 'package:ggpubr':
##
## get_legend## The following object is masked from 'package:lubridate':
##
## stamp# Definir los colores personalizados
custom_colors <- c("Control" = "#1f77b4", "SD" = "#ff7f0e")
# Crear un tema común para asegurarse de que todos los gráficos tengan el mismo tamaño de ejes
theme_custom <- theme_pubr() +
theme(axis.title.x = element_blank(), axis.title.y = element_blank(),
axis.text = element_text(size = 10), legend.position = "none")
# Crear las gráficas de boxplot para cada variable, agrupadas por Grup
p1 <- ggplot(data, aes(x = Grup, y = `HBA1c`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 15) +
labs(title = "HBA1c", y = "HBA1c (%)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 20) + # Asegurar la misma escala para el eje y
theme_custom
p2 <- ggplot(data, aes(x = Grup, y = `Acido Urico (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 15) +
labs(title = "Ácido Úrico", y = "Ácido Úrico (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 20) + # Asegurar la misma escala para el eje y
theme_custom
p3 <- ggplot(data, aes(x = Grup, y = `Glucosa (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 500) +
labs(title = "Glucosa", y = "Glucosa (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 600) + # Asegurar la misma escala para el eje y
theme_custom
p4 <- ggplot(data, aes(x = Grup, y = `Urea (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 200) +
labs(title = "Urea", y = "Urea (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 250) + # Asegurar la misma escala para el eje y
theme_custom
p5 <- ggplot(data, aes(x = Grup, y = `Creatinina (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 10) +
labs(title = "Creatinina", y = "Creatinina (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 15) + # Asegurar la misma escala para el eje y
theme_custom
# Crear la leyenda personalizada con el título "Grup"
legend <- get_legend(
ggplot(data, aes(x = Grup, y = `Creatinina (mg/dL)`, fill = Grup)) +
geom_boxplot() +
labs(fill = "Grup") + # Cambiar la etiqueta de la leyenda aquí
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
theme(legend.position = "bottom")
)## Warning in get_plot_component(plot, "guide-box"): Multiple components found;
## returning the first one. To return all, use `return_all = TRUE`.# Ajustar las proporciones relativas de los gráficos para garantizar el mismo tamaño de las celdas
top_row <- plot_grid(p1, p2, labels = c('A', 'B'), label_size = 12, ncol = 2, rel_widths = c(1, 1))## Warning: Removed 4 rows containing non-finite outside the scale range
## (`stat_boxplot()`).## Warning: Removed 4 rows containing non-finite outside the scale range
## (`stat_compare_means()`).## Warning: Removed 4 rows containing missing values or values outside the scale range
## (`geom_point()`).bottom_row <- plot_grid(p3, p4, p5, labels = c('C', 'D', 'E'), label_size = 12, ncol = 3, rel_widths = c(1, 1, 1))
# Combinar ambas filas y agregar la leyenda
combined_plot <- plot_grid(top_row, bottom_row, legend, ncol = 1, rel_heights = c(1, 1, 0.2))
# Mostrar el gráfico
print(combined_plot)
summary_table <- data %>%
select(Grup, `Colesterol (mg/dL)`, `Trigliceridos (mg/dL)`, `C - HDL (mg/dL)`,`C-LDL (mg/dL)`,`C -VLDL (mg/dL)`) %>%
tbl_summary(
by =Grup, # Agrupar por la variable dependiente "Sindrome Down"
missing = "ifany", # Excluye valores faltantes
statistic = list(
all_continuous() ~ "{median} ({IQR})"
)) %>% # Agregar el IQR a las estadísticas %>% add_overall() %>%
add_p() %>% add_overall() %>%
modify_header(label = "**Variables**") %>%
modify_caption("**Table: Laboratorios química Sanguínea de pacientes con SD**") %>%
as_gt() %>%
gt::tab_style(
style = gt::cell_text(weight = "bold"),
locations = gt::cells_column_labels()
)
summary_table| Variables | Overall N = 1051 |
CTRL N = 551 |
SD N = 501 |
p-value2 |
|---|---|---|---|---|
| Colesterol (mg/dL) | 157 (48) | 158 (44) | 154 (44) | 0.4 |
| Trigliceridos (mg/dL) | 76 (41) | 63 (22) | 97 (33) | <0.001 |
| C - HDL (mg/dL) | 53 (21) | 58 (17) | 45 (16) | <0.001 |
| C-LDL (mg/dL) | 88 (32) | 88 (31) | 87 (34) | >0.9 |
| C -VLDL (mg/dL) | 15 (8) | 13 (4) | 19 (6) | <0.001 |
| 1 Median (IQR) | ||||
| 2 Wilcoxon rank sum test | ||||
library(ggpubr)
library(cowplot)
# Definir los colores personalizados
custom_colors <- c("Control" = "#1f77b4", "SD" = "#ff7f0e")
# Crear un tema común para asegurarse de que todos los gráficos tengan el mismo tamaño de ejes
theme_custom <- theme_classic2() +
theme(axis.title.x = element_blank(), axis.title.y = element_blank(),
axis.text = element_text(size = 12), legend.position = "none")
# Crear las gráficas de boxplot para cada variable, agrupadas por Grup
p1 <- ggplot(data, aes(x = Grup, y = `Colesterol (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 400) +
labs(title = "Colesterol", y = "Colesterol (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 600) + # Asegurar la misma escala para el eje y
theme_custom
p2 <- ggplot(data, aes(x = Grup, y = `Trigliceridos (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 400) +
labs(title = "Triglicéridos", y = "Triglicéridos (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 600) + # Asegurar la misma escala para el eje y
theme_custom
p3 <- ggplot(data, aes(x = Grup, y = `C - HDL (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 150) +
labs(title = "C - HDL", y = "C - HDL (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 200) + # Asegurar la misma escala para el eje y
theme_custom
p4 <- ggplot(data, aes(x = Grup, y = `C-LDL (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 500) +
labs(title = "C - LDL", y = "C - LDL (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 600) + # Asegurar la misma escala para el eje y
theme_custom
p5 <- ggplot(data, aes(x = Grup, y = `C -VLDL (mg/dL)`, fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 200) +
labs(title = "C - VLDL", y = "C - VLDL (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(0, 300) + # Asegurar la misma escala para el eje y
theme_custom
# Crear la leyenda personalizada con el título "Grup"
legend <- get_legend(
ggplot(data, aes(x = Grup, y = `C -VLDL (mg/dL)`, fill = Grup)) +
geom_boxplot() +
labs(fill = "Grup") + # Cambiar la etiqueta de la leyenda aquí
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
theme(legend.position = "bottom")
)## Warning in get_plot_component(plot, "guide-box"): Multiple components found;
## returning the first one. To return all, use `return_all = TRUE`.# Ajustar las proporciones relativas de los gráficos para garantizar el mismo tamaño de las celdas
top_row <- plot_grid(p1, p2, labels = c('A', 'B'), label_size = 12, ncol = 2, rel_widths = c(1, 1))
bottom_row <- plot_grid(p3, p4, p5, labels = c('C', 'D', 'E'), label_size = 12, ncol = 3, rel_widths = c(1, 1, 1))
# Combinar ambas filas y agregar la leyenda
combined_plot <- plot_grid(top_row, bottom_row, legend, ncol = 1, rel_heights = c(1, 1, 0.2))
# Mostrar el gráfico
print(combined_plot)
## RAGE SOLUBLE
ggplot(data, aes(x = Grup, y = `niveles de AGEs circulantes Final` , fill = Grup)) +
geom_boxplot(outlier.shape = NA) +
geom_jitter(width = 0.2, alpha = 0.5, color = "black") +
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, label.y = 200) +
labs(title = "AGEs circulante", y = "C - VLDL (mg/dL)") +
scale_x_discrete(labels = c("Control", "SD")) +
scale_fill_manual(values = custom_colors) +
ylim(-1000, 500) + # Asegurar la misma escala para el eje y
theme_custom## Warning: Removed 7 rows containing non-finite outside the scale range
## (`stat_boxplot()`).## Warning: Removed 7 rows containing non-finite outside the scale range
## (`stat_compare_means()`).## Warning: Removed 7 rows containing missing values or values outside the scale range
## (`geom_point()`).
summary_table <- data %>%
select(Grup, `niveles de AGEs circulantes Final`) %>%
tbl_summary(
by =Grup, # Agrupar por la variable dependiente "Sindrome Down"
missing = "ifany", # Excluye valores faltantes
statistic = list(
all_continuous() ~ "{median} ({IQR})"
)) %>% # Agregar el IQR a las estadísticas %>% add_overall() %>%
add_p() %>% add_overall() %>%
modify_header(label = "**Variables**") %>%
modify_caption("**Tabla: niveles de AGEs circulantes de pacientes con SD y control**") %>%
as_gt() %>%
gt::tab_style(
style = gt::cell_text(weight = "bold"),
locations = gt::cells_column_labels()
)Polymorphism association
# Install and load SNPassoc
#install.packages("SNPassoc")
library(SNPassoc)## Registered S3 method overwritten by 'SNPassoc':
## method from
## summary.haplo.glm haplo.stats# Assuming your data is in a data frame called 'genotype_data'
# and you have loaded the data into R, with the genotype columns and Renal_Dysfunction status
# Example data loading
# genotype_data <- read.csv("your_data_file.csv")
str(data)## 'data.frame': 105 obs. of 34 variables:
## $ # : num 1 2 3 4 5 6 7 8 9 10 ...
## $ Grup : chr "SD" "SD" "SD" "SD" ...
## $ Registro : chr "SD001" "SD004" "SD005" "SD006" ...
## $ Nombre : chr "Rayli Emmanuel Caballero Gámez" "Ilda Elizabeth Hernandez Monero" "Jose Alberto Camacho Alvarez" "Charbet Valeria Valenzuela Requeña" ...
## $ Sexo : chr "Másculino" "Femenino" "Másculino" "Femenino" ...
## $ Edad : num 11 4 14 13 16 11 11 4 10 11 ...
## $ Peso (kg) : num 32.6 16.4 64.5 72.5 66 27.5 46.5 15 34 31.6 ...
## $ Talla (M) : num 1.52 0.949 1.495 1.403 1.444 ...
## $ Cintura (cm) : num 62 57.5 93 91.1 91 60.2 74.5 52.5 74.1 64.5 ...
## $ Cadera (cm) : num 72.5 58 93.5 110.5 102 ...
## $ ICC (calculada) : num 0.855 0.991 0.995 0.824 0.892 ...
## $ IMC : num 14.1 18.2 28.9 36.8 31.7 ...
## $ HBA1c : num 5.2 5.08 6.54 5.7 5.85 5.81 5.58 5.39 5.93 5.59 ...
## $ Acido Urico (mg/dL) : num NA 4.1 7.3 6.3 NA NA NA 5 5 4.8 ...
## $ Glucosa (mg/dL) : num 88 72 100 85 98 101 101 90 99 91 ...
## $ Urea (mg/dL) : num 26.6 34.3 31.1 28 35.1 36.5 20.4 21.3 41.3 28.3 ...
## $ Creatinina (mg/dL) : num 0.6 0.3 0.8 0.8 0.6 0.5 0.5 0.4 0.5 0.4 ...
## $ Colesterol (mg/dL) : num 115 153 171 163 134 138 110 135 196 131 ...
## $ Trigliceridos (mg/dL) : num 47 79 91 142 72 30 104 54 137 86 ...
## $ C - HDL (mg/dL) : num 44 44 76 55 46 72 28 31 52 63 ...
## $ C -VLDL (mg/dL) : num 9.4 15.8 18.2 48.4 14.4 6 20.8 10.8 27.4 17.2 ...
## $ C-LDL (mg/dL) : num 61.6 93.2 76.8 79.6 73.6 ...
## $ CG : num 278.3 270.8 164.3 99.6 143.5 ...
## $ 260/230 : num 1.91 1.83 1.7 1.83 1.8 1.84 1.62 1.26 1.83 1.64 ...
## $ 260/280 : num 2 1.51 0.89 1.05 1.1 1.19 1.85 0.84 2.6 1.75 ...
## $ Genotipo RAGE 625 : chr "A/A" "A/A" "A/A" "A/A" ...
## $ Genotipo RAGE 624 : chr "A/A" "A/A" "A/A" "A/T" ...
## $ Cuantificacion de AGEs fluorecentes (370nm/455nm) : num 204 183 174 249 228 ...
## $ Cuantificacion de AGEs fluorecentes (485nm/535nm) : num 218 178 171 246 218 ...
## $ Cuantificacion de proteina total : num 0.665 0.595 0.748 0.695 0.789 ...
## $ cantidad de proteina total ug/mL usando la formula de la recta : num 1304 1129 1511 1378 1613 ...
## $ cantidad de proteina total ug/mL usando la formula de la recta entre 1000: num 1.3 1.13 1.51 1.38 1.61 ...
## $ niveles de AGEs circulantes Final : num 156 162 115 181 142 ...
## $ Promedio Srage final : num 1865 2142 1856 1093 888 ...# Convert the columns to factors
data$`Genotipo RAGE 624` <- as.factor(data$`Genotipo RAGE 624`)
data$`Genotipo RAGE 625` <- as.factor(data$`Genotipo RAGE 625`)
data$Grup <- factor(data$Grup, levels = c("CTRL", "SD"))
# Hay Nas, en los SNP hay que eliminarlos
#Remove NAs
#data <- na.omit(data)
# Alternatively, if using column indices
# You can also identify the indices of these columns manually if needed:
# For example:
snp_indices <- match(c("Genotipo RAGE 625", "Genotipo RAGE 624"), colnames(data))
# Then use these indices in setupSNP
SNPs <- setupSNP(data, colSNPs = snp_indices, sep = "/")Visualize SNP summary
plot(SNPs$Genotipo.RAGE.624)
plot(SNPs$Genotipo.RAGE.625)
# You can also change the plot type to pie chart
plot(SNPs$Genotipo.RAGE.624, type = pie)
plot(SNPs$Genotipo.RAGE.625, type = pie)
### Perform association analysis for each SNP
RAGE 624
association(Grup ~ Genotipo.RAGE.624, data = SNPs)##
## SNP: Genotipo.RAGE.624 adjusted by:
## CTRL % SD % OR lower upper p-value AIC
## Codominant
## A/A 28 50.9 27 54.0 1.00 0.9224 151.2
## A/T 23 41.8 19 38.0 0.86 0.38 1.92
## T/T 4 7.3 4 8.0 1.04 0.24 4.57
## Dominant
## A/A 28 50.9 27 54.0 1.00 0.7514 149.2
## A/T-T/T 27 49.1 23 46.0 0.88 0.41 1.90
## Recessive
## A/A-A/T 51 92.7 46 92.0 1.00 0.8885 149.3
## T/T 4 7.3 4 8.0 1.11 0.26 4.69
## Overdominant
## A/A-T/T 32 58.2 31 62.0 1.00 0.6899 149.2
## A/T 23 41.8 19 38.0 0.85 0.39 1.87
## log-Additive
## 0,1,2 55 52.4 50 47.6 0.94 0.51 1.73 0.8482 149.3RAGE 625
# Perform association analysis for one SNP
association(Grup ~ Genotipo.RAGE.625, data = SNPs)##
## SNP: Genotipo.RAGE.625 adjusted by:
## CTRL % SD % OR lower upper p-value AIC
## Codominant
## A/A 45 81.8 32 64.0 1.00 0.05481 145.2
## A/G 10 18.2 16 32.0 2.25 0.90 5.59
## G/G 0 0.0 2 4.0 0.00
## Dominant
## A/A 45 81.8 32 64.0 1.00 0.03845 145.0
## A/G-G/G 10 18.2 18 36.0 2.53 1.03 6.20
## Recessive
## A/A-A/G 55 100.0 48 96.0 1.00 0.22436 146.3
## G/G 0 0.0 2 4.0 0.00
## Overdominant
## A/A-G/G 45 81.8 34 68.0 1.00 0.10062 146.6
## A/G 10 18.2 16 32.0 2.12 0.85 5.24
## log-Additive
## 0,1,2 55 52.4 50 47.6 2.58 1.11 5.99 0.05481 144.1HWE
summary(SNPs, print=FALSE)## alleles major.allele.freq HWE missing (%)
## Genotipo.RAGE.625 A/G 85.7 1 0
## Genotipo.RAGE.624 A/T 72.4 1 0gene-gene interactions
association(Grup ~ dominant(Genotipo.RAGE.625), data = SNPs)##
## SNP: dominant(Genotipo.RAGE.625 adjusted by:
## CTRL % SD % OR lower upper p-value AIC
## Codominant
## A/A 45 81.8 32 64.0 1.00 0.03845 145
## A/G-G/G 10 18.2 18 36.0 2.53 1.03 6.2
## log-Additive
## 0,1,2 55 52.4 50 47.6 2.53 1.03 6.2 145association(Grup ~ Genotipo.RAGE.625 + Sexo, data = SNPs)## Warning in terms.formula(formula, data = data): 'varlist' has changed (from
## nvar=2) to new 3 after EncodeVars() -- should no longer happen!
## Warning in terms.formula(formula, data = data): 'varlist' has changed (from
## nvar=2) to new 3 after EncodeVars() -- should no longer happen!
## Warning in terms.formula(formula, data = data): 'varlist' has changed (from
## nvar=2) to new 3 after EncodeVars() -- should no longer happen!
## Warning in terms.formula(formula, data = data): 'varlist' has changed (from
## nvar=2) to new 3 after EncodeVars() -- should no longer happen!##
## SNP: Genotipo.RAGE.625 adjusted by: Sexo
## CTRL % SD % OR lower upper p-value AIC
## Codominant
## A/A 45 81.8 32 64.0 1.00 0.04528 146.9
## A/G 10 18.2 16 32.0 2.27 0.91 5.64
## G/G 0 0.0 2 4.0 0.00
## Dominant
## A/A 45 81.8 32 64.0 1.00 0.03735 146.7
## A/G-G/G 10 18.2 18 36.0 2.55 1.04 6.26
## Recessive
## A/A-A/G 55 100.0 48 96.0 1.00 0.08213 148.0
## G/G 0 0.0 2 4.0 0.00
## Overdominant
## A/A-G/G 45 81.8 34 68.0 1.00 0.09842 148.3
## A/G 10 18.2 16 32.0 2.13 0.86 5.29
## log-Additive
## 0,1,2 55 52.4 50 47.6 2.60 1.12 6.04 0.02107 145.7glimpse(SNPs)## Rows: 105
## Columns: 34
## $ `#` <dbl> …
## $ Grup <fct> …
## $ Registro <chr> …
## $ Nombre <chr> …
## $ Sexo <chr> …
## $ Edad <dbl> …
## $ `Peso (kg)` <dbl> …
## $ `Talla (M)` <dbl> …
## $ `Cintura (cm)` <dbl> …
## $ `Cadera (cm)` <dbl> …
## $ `ICC (calculada)` <dbl> …
## $ IMC <dbl> …
## $ HBA1c <dbl> …
## $ `Acido Urico (mg/dL)` <dbl> …
## $ `Glucosa (mg/dL)` <dbl> …
## $ `Urea (mg/dL)` <dbl> …
## $ `Creatinina (mg/dL)` <dbl> …
## $ `Colesterol (mg/dL)` <dbl> …
## $ `Trigliceridos (mg/dL)` <dbl> …
## $ `C - HDL (mg/dL)` <dbl> …
## $ `C -VLDL (mg/dL)` <dbl> …
## $ `C-LDL (mg/dL)` <dbl> …
## $ CG <dbl> …
## $ `260/230` <dbl> …
## $ `260/280` <dbl> …
## $ `Cuantificacion de AGEs fluorecentes (370nm/455nm)` <dbl> …
## $ `Cuantificacion de AGEs fluorecentes (485nm/535nm)` <dbl> …
## $ `Cuantificacion de proteina total` <dbl> …
## $ `cantidad de proteina total ug/mL usando la formula de la recta` <dbl> …
## $ `cantidad de proteina total ug/mL usando la formula de la recta entre 1000` <dbl> …
## $ `niveles de AGEs circulantes Final` <dbl> …
## $ `Promedio Srage final` <dbl> …
## $ Genotipo.RAGE.625 <fct> …
## $ Genotipo.RAGE.624 <fct> …association(`niveles de AGEs circulantes Final` ~ Genotipo.RAGE.624, data = SNPs)##
## SNP: Genotipo.RAGE.624 adjusted by:
## n me se dif lower upper p-value AIC
## Codominant
## A/A 55 -257.4 65.76 0.00 0.6345 1622
## A/T 42 -325.4 78.96 -67.96 -282.6 146.70
## T/T 8 -429.8 315.08 -172.41 -568.8 223.98
## Dominant
## A/A 55 -257.4 65.76 0.00 0.4177 1620
## A/T-T/T 50 -342.1 81.74 -84.67 -288.6 119.28
## Recessive
## A/A-A/T 97 -286.9 50.44 0.00 0.4674 1620
## T/T 8 -429.8 315.08 -142.98 -527.2 241.20
## Overdominant
## A/A-T/T 63 -279.3 69.02 0.00 0.6657 1621
## A/T 42 -325.4 78.96 -46.07 -254.5 162.33
## log-Additive
## 0,1,2 -78.03 -239.0 82.91 0.3442 1620association(`niveles de AGEs circulantes Final` ~ Genotipo.RAGE.625, data = SNPs)##
## SNP: Genotipo.RAGE.625 adjusted by:
## n me se dif lower upper p-value AIC
## Codominant
## A/A 77 -298.6 46.54 0.000 0.5827 1622
## A/G 26 -324.5 158.91 -25.866 -263.3 211.5
## G/G 2 83.7 15.29 382.325 -367.3 1132.0
## Dominant
## A/A 77 -298.6 46.54 0.000 0.9778 1621
## A/G-G/G 28 -295.3 148.73 3.291 -227.8 234.4
## Recessive
## A/A-A/G 103 -305.2 52.63 0.000 0.3079 1620
## G/G 2 83.7 15.29 388.854 -354.9 1132.7
## Overdominant
## A/A-G/G 79 -288.9 45.86 0.000 0.7691 1621
## A/G 26 -324.5 158.91 -35.545 -272.2 201.1
## log-Additive
## 0,1,2 32.659 -174.9 240.2 0.7584 1621# Cargar las librerías necesarias
library(ggplot2)
library(ggpubr)
library(dplyr)
# Crear un tema común para asegurarse de que todos los gráficos tengan el mismo tamaño de ejes
theme_custom1 <- theme_classic2() + labs(x = "Genotipo", y = "Niveles de AGEs circulantes Final") +
theme(axis.title.x = element_blank(), axis.title.y = element_blank(),
axis.text = element_text(size = 12), legend.position = "none")
# Supongamos que tus datos están en un dataframe llamado `SNPs`
# Crear una nueva variable para el modelo dominante
SNPsSD <- SNPs %>% filter(Grup == "SD") %>%
mutate(Genotipo_Dominante = ifelse(`Genotipo.RAGE.625` == "A/A", "A/A", "A/G-G/G"))
# Definir los colores personalizados
custom_colors <- c("A/A" = "#1f77b4", "A/G-G/G" = "#ff7f0e")
# Crear el boxplot
boxplot_ages <- ggplot(SNPsSD, aes(x = Genotipo_Dominante, y = `niveles de AGEs circulantes Final`, fill = Genotipo_Dominante)) +
geom_boxplot(outlier.shape = NA) + # Sin mostrar outliers para mayor claridad
geom_jitter(width = 0.2, alpha = 0.5) + # Agregar puntos para cada muestra
scale_fill_manual(values = custom_colors) + # Aplicar los colores personalizados
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, method = "t.test") + # Comparación de medias con t-test Etiquetas de los ejes
theme_custom + # Aplicar el tema personalizado
theme(axis.text.x = element_text(size = 12)) # Ajustar tamaño del texto en eje X
# Mostrar el gráfico
print(boxplot_ages)
boxplot_agesprom <- ggplot(SNPsSD, aes(x = Genotipo_Dominante, y = `Promedio Srage final`, fill = Genotipo_Dominante)) +
geom_boxplot(outlier.shape = NA) + # Sin mostrar outliers para mayor claridad
geom_jitter(width = 0.2, alpha = 0.5) + # Agregar puntos para cada muestra
scale_fill_manual(values = custom_colors) + # Aplicar los colores personalizados
stat_compare_means(aes(label = ..p.signif..), label.x = 1.5, method = "t.test") + # Comparación de medias con t-tes# Etiquetas de los ejes
theme_custom + # Aplicar el tema personalizado
theme(axis.text.x = element_text(size = 12))+ # Ajustar tamaño del texto en eje X
labs(x = "Genotipo", y = "Niveles de AGEs promedio")
print(boxplot_agesprom)
plot_grid(boxplot_ages,boxplot_agesprom, labels = c('A', 'B'), label_size = 12, ncol = 2, rel_widths = c(1, 1))
#A=`niveles de AGEs circulantes Final y B=`Promedio Srage final` summary_table <- SNPsSD %>%
select(Genotipo_Dominante,`Promedio Srage final`, `niveles de AGEs circulantes Final`) %>%
tbl_summary(
by =Genotipo_Dominante, # Agrupar por la variable dependiente "Sindrome Down"
missing = "ifany", # Excluye valores faltantes
statistic = list(
all_continuous() ~ "{median} ({IQR})"
)) %>% # Agregar el IQR a las estadísticas %>% add_overall() %>%
add_p() %>% add_overall() %>%
modify_header(label = "**Variables**") %>%
modify_caption("**Tabla: niveles de AGEs circulantes y Niveles de AGEs promedio de pacientes SD polimorfismo RAGE 625**") %>%
as_gt() %>%
gt::tab_style(
style = gt::cell_text(weight = "bold"),
locations = gt::cells_column_labels()
)
summary_table| Variables | Overall N = 501 |
A/A N = 321 |
A/G-G/G N = 181 |
p-value2 |
|---|---|---|---|---|
| Promedio Srage final | 830 (540) | 867 (412) | 717 (594) | 0.5 |
| niveles de AGEs circulantes Final | 137 (55) | 129 (39) | 144 (70) | 0.4 |
| 1 Median (IQR) | ||||
| 2 Wilcoxon rank sum exact test | ||||
summary_table <- SNPsSD %>%
select(Genotipo_Dominante,`Promedio Srage final`, `niveles de AGEs circulantes Final`,`Colesterol (mg/dL)`, `Trigliceridos (mg/dL)`, `C - HDL (mg/dL)`,`C-LDL (mg/dL)`,`C -VLDL (mg/dL)`,HBA1c,`Acido Urico (mg/dL)`, `Glucosa (mg/dL)`, `Urea (mg/dL)`) %>%
tbl_summary(
by =Genotipo_Dominante, # Agrupar por la variable dependiente "Sindrome Down"
missing = "ifany", # Excluye valores faltantes
statistic = list(
all_continuous() ~ "{median} ({IQR})"
)) %>% # Agregar el IQR a las estadísticas %>% add_overall() %>%
add_p() %>% add_overall() %>%
modify_header(label = "**Variables**") %>%
modify_caption("**Tabla: niveles de AGEs circulantes de pacientes con SD polimorfismo RAGE 625**") %>%
as_gt() %>%
gt::tab_style(
style = gt::cell_text(weight = "bold"),
locations = gt::cells_column_labels()
)## The following warnings were returned during `as_gt()`:
## ! For variable `Acido Urico (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact p-value with ties
## ! For variable `Acido Urico (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact confidence intervals with ties
## ! For variable `C - HDL (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact p-value with ties
## ! For variable `C - HDL (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact confidence intervals with ties
## ! For variable `C -VLDL (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact p-value with ties
## ! For variable `C -VLDL (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact confidence intervals with ties
## ! For variable `C-LDL (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact p-value with ties
## ! For variable `C-LDL (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact confidence intervals with ties
## ! For variable `Colesterol (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact p-value with ties
## ! For variable `Colesterol (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact confidence intervals with ties
## ! For variable `Glucosa (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact p-value with ties
## ! For variable `Glucosa (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact confidence intervals with ties
## ! For variable `HBA1c` (`Genotipo_Dominante`) and "estimate", "statistic",
## "p.value", "conf.low", and "conf.high" statistics: cannot compute exact
## p-value with ties
## ! For variable `HBA1c` (`Genotipo_Dominante`) and "estimate", "statistic",
## "p.value", "conf.low", and "conf.high" statistics: cannot compute exact
## confidence intervals with ties
## ! For variable `Trigliceridos (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact p-value with ties
## ! For variable `Trigliceridos (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact confidence intervals with ties
## ! For variable `Urea (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact p-value with ties
## ! For variable `Urea (mg/dL)` (`Genotipo_Dominante`) and "estimate",
## "statistic", "p.value", "conf.low", and "conf.high" statistics: cannot
## compute exact confidence intervals with tiessummary_table| Variables | Overall N = 501 |
A/A N = 321 |
A/G-G/G N = 181 |
p-value2 |
|---|---|---|---|---|
| Promedio Srage final | 830 (540) | 867 (412) | 717 (594) | 0.5 |
| niveles de AGEs circulantes Final | 137 (55) | 129 (39) | 144 (70) | 0.4 |
| Colesterol (mg/dL) | 154 (44) | 150 (38) | 172 (52) | 0.041 |
| Trigliceridos (mg/dL) | 97 (33) | 91 (28) | 104 (47) | 0.2 |
| C - HDL (mg/dL) | 45 (16) | 43 (11) | 53 (15) | 0.035 |
| C-LDL (mg/dL) | 87 (34) | 83 (22) | 98 (41) | 0.3 |
| C -VLDL (mg/dL) | 19 (6) | 18 (6) | 20 (9) | 0.2 |
| HBA1c | 5.06 (0.77) | 5.20 (0.71) | 5.00 (0.65) | 0.11 |
| Acido Urico (mg/dL) | 5.35 (1.30) | 5.00 (1.33) | 5.40 (0.82) | 0.4 |
| Unknown | 4 | 3 | 1 | |
| Glucosa (mg/dL) | 92 (13) | 92 (9) | 90 (17) | 0.3 |
| Urea (mg/dL) | 28 (8) | 28 (6) | 25 (14) | 0.7 |
| 1 Median (IQR) | ||||
| 2 Wilcoxon rank sum exact test; Wilcoxon rank sum test | ||||
library(gtsummary)
# Convertir los elementos de la variable Genotipo_Dominante en SNPsSD
SNPsSDlr <- SNPsSD %>%
mutate(Genotipo_Dominante = recode(Genotipo_Dominante, "A/A" = 0, "A/G-G/G" = 1))
# build logistic regression model
m1 <- glm(Genotipo_Dominante ~ Sexo+Edad+IMC+`Colesterol (mg/dL)`++`C - HDL (mg/dL)`+`niveles de AGEs circulantes Final` +`Promedio Srage final`, data=SNPsSDlr, family = binomial)
print(m1)##
## Call: glm(formula = Genotipo_Dominante ~ Sexo + Edad + IMC + `Colesterol (mg/dL)` +
## +`C - HDL (mg/dL)` + `niveles de AGEs circulantes Final` +
## `Promedio Srage final`, family = binomial, data = SNPsSDlr)
##
## Coefficients:
## (Intercept) SexoMásculino
## -7.193459 -0.478843
## Edad IMC
## -0.069513 -0.075266
## `Colesterol (mg/dL)` `C - HDL (mg/dL)`
## 0.025504 0.047231
## `niveles de AGEs circulantes Final` `Promedio Srage final`
## 0.021243 -0.000341
##
## Degrees of Freedom: 49 Total (i.e. Null); 42 Residual
## Null Deviance: 65.34
## Residual Deviance: 51.48 AIC: 67.48# view raw model results
summary(m1)$coefficients## Estimate Std. Error z value
## (Intercept) -7.1934593977 3.545547495 -2.0288713
## SexoMásculino -0.4788431071 0.741335229 -0.6459198
## Edad -0.0695131092 0.125331829 -0.5546325
## IMC -0.0752663171 0.098813889 -0.7616977
## `Colesterol (mg/dL)` 0.0255041111 0.014128927 1.8050989
## `C - HDL (mg/dL)` 0.0472305196 0.028061855 1.6830862
## `niveles de AGEs circulantes Final` 0.0212425048 0.010771320 1.9721356
## `Promedio Srage final` -0.0003410134 0.000831004 -0.4103631
## Pr(>|z|)
## (Intercept) 0.04247141
## SexoMásculino 0.51833129
## Edad 0.57914604
## IMC 0.44624042
## `Colesterol (mg/dL)` 0.07105921
## `C - HDL (mg/dL)` 0.09235841
## `niveles de AGEs circulantes Final` 0.04859413
## `Promedio Srage final` 0.68153959tbl_regression(m1, exponentiate = TRUE)| Characteristic | OR1 | 95% CI1 | p-value |
|---|---|---|---|
| Sexo | |||
| Femenino | — | — | |
| Másculino | 0.62 | 0.14, 2.63 | 0.5 |
| Edad | 0.93 | 0.71, 1.18 | 0.6 |
| IMC | 0.93 | 0.75, 1.12 | 0.4 |
| Colesterol (mg/dL) | 1.03 | 1.00, 1.06 | 0.071 |
| C - HDL (mg/dL) | 1.05 | 0.99, 1.11 | 0.092 |
| niveles de AGEs circulantes Final | 1.02 | 1.00, 1.05 | 0.049 |
| Promedio Srage final | 1.00 | 1.00, 1.00 | 0.7 |
| 1 OR = Odds Ratio, CI = Confidence Interval | |||
# Cargar las librerías necesarias
library(pROC)## Type 'citation("pROC")' for a citation.##
## Attaching package: 'pROC'## The following objects are masked from 'package:stats':
##
## cov, smooth, varlibrary(dplyr)
library(ggplot2)
# Asumiendo que SNPsSD es tu dataframe y que tienes una variable "outcome" que es el resultado a predecir (0 o 1)
# Aquí "outcome" debería ser tu variable dependiente o de respuesta.
# Crear el objeto ROC
roc_obj <- roc(SNPsSDlr$Genotipo_Dominante, SNPsSDlr$`niveles de AGEs circulantes Final`)## Setting levels: control = 0, case = 1## Setting direction: controls < cases# Calcular el punto de corte óptimo usando el índice de Youden
optimal_cutoff <- coords(roc_obj, "best", ret = "threshold", best.method = "youden")
# Calcular el AUC, IC 95%, sensibilidad y especificidad
auc_value <- auc(roc_obj)
ci_auc <- ci.auc(roc_obj)
# Usar coords para extraer la sensibilidad y especificidad en el punto de corte óptimo
optimal_coords <- coords(roc_obj, x = "best", best.method = "youden", ret = c("sensitivity", "specificity"))
sensitivity <- optimal_coords[1]
specificity <- optimal_coords[2]
# Crear la curva ROC con ajustes en la posición de las anotaciones
roc_plot <- ggroc(roc_obj) +
geom_abline(slope = 1, intercept = 1, linetype = "dashed", color = "gray") +
theme_minimal() +
ggtitle("Curva ROC para Genotipo mutante A/G G/G ") +
xlab("1 - Especificidad") +
ylab("Sensibilidad")
# Mostrar la curva ROC
print(roc_plot)