summaryrefslogtreecommitdiff
path: root/R/heatmap.R
blob: cb6c8f8fb2c78630ead946416c68bbdb1caa638b (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
#' Heatmap Plot (Upper Triangle)
#'
#' Minimal ggplot2 relationship structure plot showing the upper triangle
#' in the top-right corner with no axis labels.
#' @param R A square numeric correlation matrix
#' @param title Optional plot title
#' @param labels Logical to plot column names or not (default FALSE)
#' @param order Optional character vector of the order you want the plot sorted in
#'
#' @return A ggplot object
#'
#' @examples
#' X <- matrix(data = rnorm(100), nrow = 10)
#' R <- cor(X)
#' heatmap(R, title = 'Example Correlation Plot')
#'
#' @importFrom reshape2 melt
#' @import ggplot2
#' @export
heatmap <- function(R, title = NULL, labels = FALSE, order = NULL){
	R_list <- tryCatch({
		if(is.matrix(R)){
			Rm <- R
			cnames <- colnames(Rm)
			rnames <- rownames(Rm)
		}else{
			Rm <- as.matrix(R)
			cnames <- colnames(Rm)
			rnames <- rownames(Rm)
		}
		if(nrow(Rm) == ncol(Rm)){
			if(!is.null(order)){
				order <- get_order(Rm)
				M <- Rm[order, order]
			}
			list(
				M = M[upper.tri(M)],
				cnames = cnames,
				rnames = rnames,
				p = ncol(M)
			)
		}else{
			stop('Input must be coercible to a square matrix')
		}

	}, error = function(e){
		stop(paste('Invalid input for heatmap:', e$message))
	})

	if(is.null(R_list$cnames)){
		R_list$cnames <- as.character(seq_len(R_list$p))
	}
	if(is.null(R_list$rnames)){
		R_list$rnames <- as.character(seq_len(R_list$p))
	}

	# Now get a df from the upper triangle
	m <- length(R_list$R)
	n <- as.integer((1 + sqrt(1 + 8 * m)) / 2)
	M <- matrix(NA, n, n)
	M[upper.tri(M)] <- R_list$R
	colnames(M) <- R_list$cnames
	rownames(M) <- R_list$rnames
	df <- reshape2::melt(M, na.rm = TRUE)
	colnames(df) <- c('i', 'j', 'value')
	df$i <- as.numeric(df$i)
	df$j <- as.numeric(df$j)

	# And plot it
	p <- ggplot2::ggplot(
		data = df,
		ggplot2::aes(x = .data$j, y = .data$i, fill = .data$value)
	) +
	ggplot2::geom_tile() +
	ggplot2::scale_y_reverse() +
	ggplot2::coord_fixed(clip = 'off') +
	ggplot2::scale_fill_gradient2(
		low = 'red',
		mid = 'white',
		high = 'blue',
		midpoint = 0,
		limits = c(-1, 1),
		name = NULL,
		guide = 'none'
	) +
	ggplot2::theme_void()

	if(!is.null(title)){
		p <- p +
		ggplot2::ggtitle(title) +
		ggplot2::theme(
			plot.title = ggplot2::element_text(hjust = 0.5)
		)
	}

	if(labels){
		col_label_df <- data.frame(
			j = seq_len(n)[-1],
			i = 0,
			label = R_list$cnames[-1]
		)
		row_label_df <- data.frame(
			j = seq_len(n)[-n],
			i = seq_len(n)[-n],
			label = R_list$rnames[-n]
		)

		max_nchar <- max(nchar(R_list$cnames))
		top_margin <- 5 + max_nchar * 2

		p <- p +
		ggplot2::geom_text(
			data = col_label_df,
			ggplot2::aes(x = .data$j, y = .data$i, label = .data$label),
			inherit.aes = FALSE,
			angle = 90,
			hjust = 0,
			vjust = 0.5
		) +
		ggplot2::geom_text(
			data = row_label_df,
			ggplot2::aes(x = .data$j, y = .data$i, label = .data$label),
			inherit.aes = FALSE,
			hjust = 1,
			vjust = 0.5
		) +
		ggplot2::theme(
			plot.margin = ggplot2::margin(
				t = top_margin,
				r = 5,
				b = 5,
				l = 5
			)
		)
	}

	return(p)
}