data <- c(rnorm(5,mean=5), rnorm(5,mean=20), rnorm(5, mean=100), c(600,700,800,900,10000))
data <- matrix(data, ncol=5, byrow=T)
data <- as.data.frame(data)
rownames(data) <- letters[1:4]
colnames(data) <- paste("Grp", 1:5, sep="_")
data
Grp_1 Grp_2 Grp_3 Grp_4 Grp_5
a 6.61047 20.946720 100.133106 600.000000 5.267921
b 20.80792 99.865962 700.000000 3.737228 19.289715
c 100.06930 800.000000 6.252753 21.464081 98.607518
d 900.00000 3.362886 20.334078 101.117728 10000.000000
# 对数转换
# +1是为了防止对0取对数;是加1还是加个更小的值取决于数据的分布。
# 加的值一般认为是检测的低阈值,低于这个值的数字之间的差异可以忽略。
data_log <- log2(data+1)
data_log
Grp_1 Grp_2 Grp_3 Grp_4 Grp_5
a 2.927986 4.455933 6.660112 9.231221 2.647987
b 4.446780 6.656296 9.453271 2.244043 4.342677
c 6.659201 9.645658 2.858529 4.489548 6.638183
d 9.815383 2.125283 4.415088 6.674090 13.287857
data_log$ID = rownames(data_log)
data_log_m = melt(data_log, id.vars=c("ID"))
p <- ggplot(data_log_m, aes(x=variable,y=ID)) + xlab("samples") + ylab(NULL) + theme_bw() + theme(panel.grid.major = element_blank()) + theme(legend.key=element_blank()) + theme(axis.text.x=element_text(angle=45,hjust=1, vjust=1)) + theme(legend.position="top") + geom_tile(aes(fill=value)) + scale_fill_gradient(low = "white", high = "red")
ggsave(p, filename="heatmap_log.pdf", width=8, height=12, units=c("cm"),colormodel="srgb")