SCI图片复现：面积图

省流：完整代码拉到最后

首先数据表格要求：

1. 趋势气泡图数据：这个数据应该是一个CSV文件，具有三个列。第一列应该是"Diameter"，代表直径，应该是数字。第二列应该是"expression"，代表表达量，也应该是数字。第三列应该是"group"，代表组，这是一个分类变量。这个CSV文件不应该有行名。

2. Nature图数据：这个数据也应该是一个CSV文件，具有三个列。第一列应该是"celltype"，代表细胞类型，这是一个分类变量。第二列应该是"stage"，代表阶段，这也是一个分类变量。第三列应该是"Freq"，代表频率，应该是数字。这个CSV文件不应该有行名。

以下是这两种数据的示例：

趋势气泡图数据：

Nature图数据：

请注意，所有的数据都应该是完整的，也就是说，没有缺失值。如果有缺失值，那么可能会导致代码运行错误。

步骤1：生成示例数据

首先，我会为这两种图形生成示例数据。

1. 对于趋势气泡图，数据可能包含以下列：“Diameter”（直径），“expression”（表达量）以及“group”（组别）。
2. 对于nature图，数据可能包含以下列：“celltype”（细胞类型），“stage”（阶段）和“Freq”（频率）。

# 创建趋势气泡图数据
set.seed(123)  # 设置随机数种子以保证结果可复现
trend_data <- data.frame(
  Diameter = sample(100:500, 200, replace = TRUE),
  expression = runif(200, 0, 100),
  group = factor(sample(LETTERS[1:4], 200, replace = TRUE))
)
head(trend_data)  # 打印前几行数据

# 创建nature数据
nature_data <- data.frame(
  celltype = factor(rep(LETTERS[1:4], each = 50)),
  stage = factor(sample(c("CS12", "CS13", "CS15", "CS17", "CS20"), 200, replace = TRUE)),
  Freq = runif(200, 0, 0.5)
)
head(nature_data)  # 打印前几行数据

修改后的趋势气泡图代码

library(ggplot2)

# 读入趋势气泡图数据
# data <- read.csv("趋势气泡图.csv", header = T)  # 原始代码
data <- trend_data  # 修改后的代码

col <- c("#D57444", "#33B1C0", "#34993B", "#8C67AC")

ggplot(data, aes(x=Diameter, y=expression, fill=group)) + 
  geom_area(size=0.01,colour="white") +  # 加面积
  labs(y="Expression") +
  theme_bw() +
  scale_fill_manual(values=col) +
  theme(panel.grid.major=element_blank(),panel.grid.minor=element_blank(),
        axis.text = element_text(color = "black",size = 10),
        axis.title = element_text(color = 'black', size=12)) +
  geom_vline(aes(xintercept =100),linetype="dashed", size=1.2, colour="white") +  # 加竖线
  geom_vline(aes(xintercept =200),linetype="dashed", size=1.2, colour="white") +
  geom_vline(aes(xintercept =300),linetype="dashed", size=1.2, colour="white") +
  geom_vline(aes(xintercept =400),linetype="dashed", size=1.2, colour="white")

修改后的Nature图代码

# 读入Nature数据
# df <- read.csv("nature_ratio.csv", header = T)  # 原始代码
df <- nature_data  # 修改后的代码

table(df$Stage)
A <- prop.table(table(df$celltype, df$stage), margin = 2)
A <- as.data.frame(A)
colnames(A) <- c("celltype", "stage", "Freq")

cluster_cols <- c("#DC050C", "#FB8072", "#1965B0", "#7BAFDE", "#882E72",
                  "#B17BA6", "#FF7F00", "#FDB462", "#E7298A", "#E78AC3",
                  "#33A02C", "#B2DF8A", "#55A1B1", "#8DD3C7", "#A6761D",
                  "#E6AB02", "#7570B3", "#BEAED4", "#666666", "#999999",
                  "#aa8282", "#d4b7b7", "#8600bf", "#ba5ce3", "#808000",
                  "#aeae5c", "#1e90ff", "#00bfff", "#56ff0d", "#ffff00")

ggplot(A,aes(x = stage, y =Freq, group=celltype)) +
  stat_summary(geom = 'line',fun='mean',cex=1,col='white') +  # 先要有折线
  geom_area(data = A,aes(fill=celltype)) +  # 折线下面积，填充用celltype
  scale_fill_manual(values=cluster_cols) +
  labs(x=NULL, y=NULL) +
  theme_bw() +
  theme(panel.grid.major=element_blank(),panel.grid.minor=element_blank(),
        axis.text = element_text(color = "black",size = 10)) +
  geom_vline(aes(xintercept ="CS12"),linetype="dashed", size=1.2, colour="white") +
  geom_vline(aes(xintercept ="CS13"),linetype="dashed", size=1.2, colour="white") +
  geom_vline(aes(xintercept ="CS15"),linetype="dashed", size=1.2, colour="white") +
  geom_vline(aes(xintercept ="CS17"),linetype="dashed", size=1.2, colour="white") +
  geom_vline(aes(xintercept ="CS20"),linetype="dashed", size=1.2, colour="white")

优化后的趋势气泡图代码

library(ggplot2)

# 读入趋势气泡图数据
data <- trend_data  # 使用我们前面创建的示例数据

col <- c("#D57444", "#33B1C0", "#34993B", "#8C67AC")  # 指定每个组别的颜色

# 创建ggplot对象
p <- ggplot(data, aes(x = Diameter, y = expression, fill = group))

# 添加图层
p <- p +
  geom_area(size = 0.01, colour = "white") +  # 添加面积层
  labs(y = "Expression") +  # 设置y轴标签
  theme_bw() +  # 使用黑白主题
  scale_fill_manual(values = col) +  # 手动设置填充颜色
  theme(
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    axis.text = element_text(color = "black", size = 10),
    axis.title = element_text(color = 'black', size = 12)
  )  # 设置主题元素

# 添加竖线
for (i in seq(100, 400, by = 100)) {
  p <- p + geom_vline(aes(xintercept = i), linetype = "dashed", size = 1.2, colour = "white")
}

p  # 打印图形

优化后的Nature图代码

# 读入Nature数据
df <- nature_data  # 使用我们前面创建的示例数据

# 计算每个阶段的频率
A <- prop.table(table(df$celltype, df$stage), margin = 2)
A <- as.data.frame(A)
colnames(A) <- c("celltype", "stage", "Freq")

# 指定每个细胞类型的颜色
cluster_cols <- c("#DC050C", "#FB8072", "#1965B0", "#7BAFDE", "#882E72",
                  "#B17BA6", "#FF7F00", "#FDB462", "#E7298A", "#E78AC3",
                  "#33A02C", "#B2DF8A", "#55A1B1", "#8DD3C7", "#A6761D",
                  "#E6AB02", "#7570B3", "#BEAED4", "#666666", "#999999",
                  "#aa8282", "#d4b7b7", "#8600bf", "#ba5ce3", "#808000",
                  "#aeae5c", "#1e90ff", "#00bfff", "#56ff0d", "#ffff00")

# 创建ggplot对象
p <- ggplot(A, aes(x = stage, y = Freq, group = celltype))

# 添加图层
p <- p +
  stat_summary(geom = 'line', fun = 'mean', cex = 1, col = 'white') +  # 添加折线
  geom_area(aes(fill = celltype)) +  # 添加面积层
  scale_fill_manual(values = cluster_cols) +  # 手动设置填充颜色
  labs(x = NULL, y = NULL) +  # 清除x和y轴的标签
  theme_bw() +  # 使用黑白主题
  theme(
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    axis.text = element_text(color = "black", size = 10)
  )  # 设置主题元素

# 添加竖线
for (stage in c("CS12", "CS13", "CS15", "CS17", "CS20")) {
  p <- p + geom_vline(aes(xintercept = stage), linetype = "dashed", size = 1.2, colour = "white")
}

p  # 打印图形

优化代码，使得图片更为丰富和美观。我将使用ggthemes包和ggtitle函数来改善图表的外观，并给图表添加标题。

# 安装和加载ggthemes包
if (!require(ggthemes)) {
  install.packages("ggthemes")
}
library(ggthemes)

# 优化趋势气泡图
p_trend <- ggplot(trend_data, aes(x = Diameter, y = expression, fill = group)) +
  geom_area(size = 0.01, colour = "white") +
  scale_fill_manual(values = col) +
  labs(x = "Diameter", y = "Expression", title = "Trend Bubble Plot") +
  theme_tufte() +  # 使用Tufte主题，这个主题具有简洁的风格
  theme(
    plot.title = element_text(hjust = 0.5),
    legend.position = "bottom",
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    axis.text = element_text(color = "black", size = 10),
    axis.title = element_text(color = 'black', size = 12)
  )
for (i in seq(100, 400, by = 100)) {
  p_trend <- p_trend + geom_vline(aes(xintercept = i), linetype = "dashed", size = 1.2, colour = "white")
}
p_trend

# 优化Nature图
p_nature <- ggplot(A, aes(x = stage, y = Freq, group = celltype)) +
  stat_summary(geom = 'line', fun = 'mean', cex = 1, col = 'white') +
  geom_area(aes(fill = celltype)) +
  scale_fill_manual(values = cluster_cols) +
  labs(x = NULL, y = NULL, title = "Nature Plot") +
  theme_tufte() +
  theme(
    plot.title = element_text(hjust = 0.5),
    legend.position = "bottom",
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    axis.text = element_text(color = "black", size = 10)
  )
for (stage in c("CS12", "CS13", "CS15", "CS17", "CS20")) {
  p_nature <- p_nature + geom_vline(aes(xintercept = stage), linetype = "dashed", size = 1.2, colour = "white")
}
p_nature

在这一步中，我们使用了ggthemes包的Tufte主题，它是一个简洁且对比度较高的主题。我们还使用了ggtitle函数来给图表添加标题，并将图例放在图表底部以更好地利用空间。另外，我们添加了panel.grid.major和panel.grid.minor的element_blank()元素，以移除主要和次要的网格线。此外，我们使用了theme函数的hjust参数来水平居中标题，并使用了theme函数的legend.position参数将图例放在底部。

完整代码

# 安装和加载需要的包
if (!require(ggplot2)) {
  install.packages("ggplot2")
}
if (!require(ggthemes)) {
  install.packages("ggthemes")
}
library(ggplot2)
library(ggthemes)

# 定义数据
trend_data <- data.frame(Diameter = c(50, 100, 150, 200, 250),
                         expression = c(0.5, 0.6, 0.7, 0.6, 0.5),
                         group = c("A", "B", "C", "A", "B"))

nature_data <- data.frame(celltype = c("Type1", "Type2", "Type3", "Type1", "Type2"),
                          stage = c("CS12", "CS13", "CS15", "CS17", "CS20"),
                          Freq = c(0.2, 0.3, 0.5, 0.3, 0.2))

# 定义颜色
col <- c("#D57444", "#33B1C0", "#34993B", "#8C67AC")
cluster_cols <- c("#DC050C", "#FB8072", "#1965B0", "#7BAFDE", "#882E72",
                  "#B17BA6", "#FF7F00", "#FDB462", "#E7298A", "#E78AC3",
                  "#33A02C", "#B2DF8A", "#55A1B1", "#8DD3C7", "#A6761D",
                  "#E6AB02", "#7570B3", "#BEAED4", "#666666", "#999999",
                  "#aa8282", "#d4b7b7", "#8600bf", "#ba5ce3", "#808000",
                  "#aeae5c", "#1e90ff", "#00bfff", "#56ff0d", "#ffff00")

# 创建趋势气泡图
p_trend <- ggplot(trend_data, aes(x = Diameter, y = expression, fill = group)) +
  geom_area(linewidth = 0.01, colour = "white") +
  scale_fill_manual(values = col) +
  labs(x = "Diameter", y = "Expression", title = "Trend Bubble Plot") +
  theme_tufte() +
  theme(
    plot.title = element_text(hjust = 0.5),
    legend.position = "bottom",
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    axis.text = element_text(color = "black", size = 10),
    axis.title = element_text(color = 'black', size = 12)
  )
for (i in seq(100, 400, by = 100)) {
  p_trend <- p_trend + geom_vline(aes(xintercept = i), linetype = "dashed", size = 1.2, colour = "white")
}

# 创建Nature图
p_nature <- ggplot(nature_data, aes(x = stage, y = Freq, group = celltype)) +
  stat_summary(geom = 'line', fun = 'mean', cex = 1, col = 'white') +
  geom_area(aes(fill = celltype)) +
  scale_fill_manual(values = cluster_cols) +
  labs(x = NULL, y = NULL, title = "Nature Plot") +
  theme_tufte() +
  theme(
    plot.title = element_text(hjust = 0.5),
    legend.position = "bottom",
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    axis.text = element_text(color = "black", size = 10)
  )
for (stage in c("CS12", "CS13", "CS15", "CS17", "CS20")) {
  p_nature <- p_nature + geom_vline(aes(xintercept = stage), linetype = "dashed", size = 1.2, colour = "white")
}

# 显示图像
print(p_trend)
print(p_nature)

# 保存图像为PDF
ggsave(filename = "TrendBubblePlot.pdf", plot = p_trend, device = "pdf", width = 10, height = 7, units = "in")
ggsave(filename = "NaturePlot.pdf", plot = p_nature, device = "pdf", width = 10, height = 7, units = "in")

# 保存数据为CSV
write.csv(trend_data, file = "TrendData.csv", row.names = FALSE)
write.csv(nature_data, file = "NatureData.csv", row.names = FALSE)