... | ... |
@@ -1142,3 +1142,15 @@ setAs("list", "HeatmapList", function(from) { |
1142 | 1142 |
} |
1143 | 1143 |
ht_list |
1144 | 1144 |
}) |
1145 |
+ |
|
1146 |
+ |
|
1147 |
+draw_heatmap_in_jupyter = function(ht, ...) { |
|
1148 |
+ width = getOption("repr.plot.width") |
|
1149 |
+ height = getOption("repr.plot.height") |
|
1150 |
+ |
|
1151 |
+ p = grid.grabExpr({ht <- draw(ht, ...)}, width = width, height = height) |
|
1152 |
+ grid.draw(p) |
|
1153 |
+ |
|
1154 |
+ invisible(ht) |
|
1155 |
+} |
|
1156 |
+ |
... | ... |
@@ -286,10 +286,10 @@ subset_gp = function(gp, i) { |
286 | 286 |
|
287 | 287 |
|
288 | 288 |
get_text_just = function(rot, side) { |
289 |
- rot = rot %% 360 |
|
290 |
- if(! rot %in% c(0, 90, 270)) { |
|
291 |
- stop_wrap("Only support horizontal or vertical rotations for text.\n") |
|
292 |
- } |
|
289 |
+ rot = rot %% 180 |
|
290 |
+ # if(! rot %in% c(0, 90, 270)) { |
|
291 |
+ # stop_wrap("Only support horizontal or vertical rotations for text.\n") |
|
292 |
+ # } |
|
293 | 293 |
if(side == "left") { |
294 | 294 |
if(rot == 0) { |
295 | 295 |
return(c(1, 0.5)) |
... | ... |
@@ -299,27 +299,27 @@ get_text_just = function(rot, side) { |
299 | 299 |
return(c(0.5, 1)) |
300 | 300 |
} |
301 | 301 |
} else if(side == "right") { |
302 |
- if(rot == 0) { |
|
302 |
+ if(rot >= 0 && rot < 90) { |
|
303 | 303 |
return(c(0, 0.5)) |
304 | 304 |
} else if(rot == 90) { |
305 | 305 |
return(c(0.5, 1)) |
306 |
- } else if(rot == 270) { |
|
307 |
- return(c(0.5, 0)) |
|
306 |
+ } else if(rot > 90 && rot < 180) { |
|
307 |
+ return(c(0, 0.5)) |
|
308 | 308 |
} |
309 | 309 |
} else if(side == "top") { |
310 | 310 |
if(rot == 0) { |
311 | 311 |
return(c(0.5, 0)) |
312 |
- } else if(rot == 90) { |
|
312 |
+ } else if(rot > 0 && rot <= 90) { |
|
313 | 313 |
return(c(0, 0.5)) |
314 |
- } else if(rot == 270) { |
|
314 |
+ } else if(rot > 90 && rot <= 180) { |
|
315 | 315 |
return(c(1, 0.5)) |
316 | 316 |
} |
317 | 317 |
} else if(side == "bottom") { |
318 | 318 |
if(rot == 0) { |
319 | 319 |
return(c(0.5, 1)) |
320 |
- } else if(rot == 90) { |
|
320 |
+ } else if(rot > 0 && rot <= 90) { |
|
321 | 321 |
return(c(1, 0.5)) |
322 |
- } else if(rot == 270) { |
|
322 |
+ } else if(rot > 90 && rot <= 180) { |
|
323 | 323 |
return(c(0, 0.5)) |
324 | 324 |
} |
325 | 325 |
} |
... | ... |
@@ -100,7 +100,7 @@ default_col = function(x, main_matrix = FALSE) { |
100 | 100 |
} else { |
101 | 101 |
if(length(unique(x)) >= 100) { |
102 | 102 |
q1 = quantile(x, 0.01, na.rm = TRUE) |
103 |
- q2 = quantile(x, 0.99, na.rm = TRUE`) |
|
103 |
+ q2 = quantile(x, 0.99, na.rm = TRUE) |
|
104 | 104 |
if(q1 == q2) { |
105 | 105 |
col_fun = colorRamp2(seq(min(x), max(x), length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
106 | 106 |
} else if(length(unique(x[x > q1 & x < q2])) == 1) { |
... | ... |
@@ -87,7 +87,7 @@ default_col = function(x, main_matrix = FALSE) { |
87 | 87 |
cat("This matrix has both negative and positive values, use a color mapping symmetric to zero\n") |
88 | 88 |
} |
89 | 89 |
if(length(unique(x)) >= 100) { |
90 |
- q1 = quantile(abs(x), 0.99) |
|
90 |
+ q1 = quantile(abs(x), 0.99, na.rm = TRUE) |
|
91 | 91 |
col_fun = colorRamp2(seq(-q1, q1, length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
92 | 92 |
|
93 | 93 |
if(any(x > q1*3 | x < -q1*3)) { |
... | ... |
@@ -99,8 +99,8 @@ default_col = function(x, main_matrix = FALSE) { |
99 | 99 |
} |
100 | 100 |
} else { |
101 | 101 |
if(length(unique(x)) >= 100) { |
102 |
- q1 = quantile(x, 0.01) |
|
103 |
- q2 = quantile(x, 0.99) |
|
102 |
+ q1 = quantile(x, 0.01, na.rm = TRUE) |
|
103 |
+ q2 = quantile(x, 0.99, na.rm = TRUE`) |
|
104 | 104 |
if(q1 == q2) { |
105 | 105 |
col_fun = colorRamp2(seq(min(x), max(x), length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
106 | 106 |
} else if(length(unique(x[x > q1 & x < q2])) == 1) { |
... | ... |
@@ -117,7 +117,8 @@ default_col = function(x, main_matrix = FALSE) { |
117 | 117 |
} |
118 | 118 |
} else { |
119 | 119 |
#col_fun = colorRamp2(range(min(x), max(x)), c("white", hsv(runif(1), 1, 1))) |
120 |
- col_fun = colorRamp2(range(min(x), max(x)), c("white", rand_color(1, luminosity = sample(c("bright", "dark"), 1)))) |
|
120 |
+ rc = rand_color(1, luminosity = sample(c("bright", "dark"), 1)) |
|
121 |
+ col_fun = colorRamp2(range(min(x), max(x)), c("white", rc)) |
|
121 | 122 |
} |
122 | 123 |
return(col_fun) |
123 | 124 |
} |
... | ... |
@@ -241,7 +241,11 @@ get_dend_order = function(x) { |
241 | 241 |
recycle_gp = function(gp, n = 1) { |
242 | 242 |
for(i in seq_along(gp)) { |
243 | 243 |
x = gp[[i]] |
244 |
- gp[[i]] = c(rep(x, floor(n/length(x))), x[seq_len(n %% length(x))]) |
|
244 |
+ if(n > 0) { |
|
245 |
+ gp[[i]] = c(rep(x, floor(n/length(x))), x[seq_len(n %% length(x))]) |
|
246 |
+ } else { |
|
247 |
+ gp[[i]] = x[1] |
|
248 |
+ } |
|
245 | 249 |
} |
246 | 250 |
return(gp) |
247 | 251 |
} |
... | ... |
@@ -88,31 +88,31 @@ default_col = function(x, main_matrix = FALSE) { |
88 | 88 |
} |
89 | 89 |
if(length(unique(x)) >= 100) { |
90 | 90 |
q1 = quantile(abs(x), 0.99) |
91 |
- col_fun = colorRamp2(c(-q1, 0, q1), c("blue", "#EEEEEE", "red")) |
|
91 |
+ col_fun = colorRamp2(seq(-q1, q1, length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
|
92 | 92 |
|
93 | 93 |
if(any(x > q1*3 | x < -q1*3)) { |
94 | 94 |
message_wrap("The automatically generated colors map from the minus and plus 99^th of the absolute values in the matrix. There are outliers in the matrix whose patterns might be hidden by this color mapping. You can manually set the color to `col` argument.\n\nUse `suppressMessages()` to turn off this message.") |
95 | 95 |
} |
96 | 96 |
} else { |
97 | 97 |
q1 = max(abs(x)) |
98 |
- col_fun = colorRamp2(c(-q1, 0, q1), c("blue", "#EEEEEE", "red")) |
|
98 |
+ col_fun = colorRamp2(seq(-q1, q1, length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
|
99 | 99 |
} |
100 | 100 |
} else { |
101 | 101 |
if(length(unique(x)) >= 100) { |
102 | 102 |
q1 = quantile(x, 0.01) |
103 | 103 |
q2 = quantile(x, 0.99) |
104 | 104 |
if(q1 == q2) { |
105 |
- col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
105 |
+ col_fun = colorRamp2(seq(min(x), max(x), length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
|
106 | 106 |
} else if(length(unique(x[x > q1 & x < q2])) == 1) { |
107 |
- col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
107 |
+ col_fun = colorRamp2(seq(min(x), max(x), length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
|
108 | 108 |
} else { |
109 |
- col_fun = colorRamp2(seq(q1, q2, length = 3), c("blue", "#EEEEEE", "red")) |
|
109 |
+ col_fun = colorRamp2(seq(q1, q2, length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
|
110 | 110 |
if(any(x > q2 + (q2-q1) | x < q1 - (q2-q1))) { |
111 | 111 |
message_wrap("The automatically generated colors map from the 1^st and 99^th of the values in the matrix. There are outliers in the matrix whose patterns might be hidden by this color mapping. You can manually set the color to `col` argument.\n\nUse `suppressMessages()` to turn off this message.") |
112 | 112 |
} |
113 | 113 |
} |
114 | 114 |
} else { |
115 |
- col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
115 |
+ col_fun = colorRamp2(seq(min(x), max(x), length.out = length(ht_opt$COLOR)), ht_opt$COLOR) |
|
116 | 116 |
} |
117 | 117 |
} |
118 | 118 |
} else { |
... | ... |
@@ -1105,6 +1105,12 @@ is_RStudio_current_dev = function() { |
1105 | 1105 |
} |
1106 | 1106 |
} |
1107 | 1107 |
|
1108 |
+is.na.expression = function(x) { |
|
1109 |
+ n = length(x) |
|
1110 |
+ sapply(seq_len(n), function(i) { |
|
1111 |
+ identical(as.character(x[i]), "__NA__") |
|
1112 |
+ }) |
|
1113 |
+} |
|
1108 | 1114 |
|
1109 | 1115 |
# dev.size = function(units = "in") { |
1110 | 1116 |
# ds = par("din") |
... | ... |
@@ -24,6 +24,14 @@ increase_heatmap_index = function() { |
24 | 24 |
INDEX_ENV$I_HEATMAP = INDEX_ENV$I_HEATMAP + 1 |
25 | 25 |
} |
26 | 26 |
|
27 |
+get_oncoprint_index = function() { |
|
28 |
+ INDEX_ENV$I_ONCOPRINT |
|
29 |
+} |
|
30 |
+ |
|
31 |
+increase_oncoprint_index = function() { |
|
32 |
+ INDEX_ENV$I_ONCOPRINT = INDEX_ENV$I_ONCOPRINT + 1 |
|
33 |
+} |
|
34 |
+ |
|
27 | 35 |
get_annotation_index = function() { |
28 | 36 |
INDEX_ENV$I_ANNOTATION |
29 | 37 |
} |
... | ... |
@@ -1097,14 +1097,14 @@ is_RStudio_current_dev = function() { |
1097 | 1097 |
} |
1098 | 1098 |
|
1099 | 1099 |
|
1100 |
-dev.size = function(units = "in") { |
|
1101 |
- ds = par("din") |
|
1100 |
+# dev.size = function(units = "in") { |
|
1101 |
+# ds = par("din") |
|
1102 | 1102 |
|
1103 |
- if(units == "cm") { |
|
1104 |
- ds = ds*2.54 |
|
1105 |
- } else if(units == "px") { |
|
1106 |
- stop("px is not supported.") |
|
1107 |
- } |
|
1103 |
+# if(units == "cm") { |
|
1104 |
+# ds = ds*2.54 |
|
1105 |
+# } else if(units == "px") { |
|
1106 |
+# stop("px is not supported.") |
|
1107 |
+# } |
|
1108 | 1108 |
|
1109 |
- ds |
|
1110 |
-} |
|
1109 |
+# ds |
|
1110 |
+# } |
... | ... |
@@ -1095,3 +1095,16 @@ is_RStudio_current_dev = function() { |
1095 | 1095 |
} |
1096 | 1096 |
} |
1097 | 1097 |
} |
1098 |
+ |
|
1099 |
+ |
|
1100 |
+dev.size = function(units = "in") { |
|
1101 |
+ ds = par("din") |
|
1102 |
+ |
|
1103 |
+ if(units == "cm") { |
|
1104 |
+ ds = ds*2.54 |
|
1105 |
+ } else if(units == "px") { |
|
1106 |
+ stop("px is not supported.") |
|
1107 |
+ } |
|
1108 |
+ |
|
1109 |
+ ds |
|
1110 |
+} |
... | ... |
@@ -239,7 +239,7 @@ recycle_gp = function(gp, n = 1) { |
239 | 239 |
|
240 | 240 |
check_gp = function(gp) { |
241 | 241 |
if(!"lineheight" %in% names(gp)) { |
242 |
- gp$lineheight = 0.8 |
|
242 |
+ gp$lineheight = 0.9 |
|
243 | 243 |
} |
244 | 244 |
if(!inherits(gp, "gpar")) { |
245 | 245 |
stop_wrap("Graphic parameters should be specified by `gpar()`.") |
... | ... |
@@ -238,6 +238,9 @@ recycle_gp = function(gp, n = 1) { |
238 | 238 |
} |
239 | 239 |
|
240 | 240 |
check_gp = function(gp) { |
241 |
+ if(!"lineheight" %in% names(gp)) { |
|
242 |
+ gp$lineheight = 0.8 |
|
243 |
+ } |
|
241 | 244 |
if(!inherits(gp, "gpar")) { |
242 | 245 |
stop_wrap("Graphic parameters should be specified by `gpar()`.") |
243 | 246 |
} |
... | ... |
@@ -92,8 +92,10 @@ default_col = function(x, main_matrix = FALSE) { |
92 | 92 |
if(length(unique(x)) >= 100) { |
93 | 93 |
q1 = quantile(x, 0.01) |
94 | 94 |
q2 = quantile(x, 0.99) |
95 |
- if(length(unique(x[x > q1 & x < q2])) == 1) { |
|
96 |
- col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
95 |
+ if(q1 == q2) { |
|
96 |
+ col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
97 |
+ } else if(length(unique(x[x > q1 & x < q2])) == 1) { |
|
98 |
+ col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
97 | 99 |
} else { |
98 | 100 |
col_fun = colorRamp2(seq(q1, q2, length = 3), c("blue", "#EEEEEE", "red")) |
99 | 101 |
if(any(x > q2 + (q2-q1) | x < q1 - (q2-q1))) { |
... | ... |
@@ -82,7 +82,7 @@ default_col = function(x, main_matrix = FALSE) { |
82 | 82 |
col_fun = colorRamp2(c(-q1, 0, q1), c("blue", "#EEEEEE", "red")) |
83 | 83 |
|
84 | 84 |
if(any(x > q1*3 | x < -q1*3)) { |
85 |
- message_wrap("The automatically generated color maps from the minus and plus 99^th of the absolute values in the matrix. There are outliers in the matrix whose patterns might be hidden by this color mapping.\n\nUse `suppressMessages()` to turn off this message.") |
|
85 |
+ message_wrap("The automatically generated colors map from the minus and plus 99^th of the absolute values in the matrix. There are outliers in the matrix whose patterns might be hidden by this color mapping. You can manually set the color to `col` argument.\n\nUse `suppressMessages()` to turn off this message.") |
|
86 | 86 |
} |
87 | 87 |
} else { |
88 | 88 |
q1 = max(abs(x)) |
... | ... |
@@ -97,7 +97,7 @@ default_col = function(x, main_matrix = FALSE) { |
97 | 97 |
} else { |
98 | 98 |
col_fun = colorRamp2(seq(q1, q2, length = 3), c("blue", "#EEEEEE", "red")) |
99 | 99 |
if(any(x > q2 + (q2-q1) | x < q1 - (q2-q1))) { |
100 |
- message_wrap("The automatically generated color maps from the 1^st and 99^th of the values in the matrix. There are outliers in the matrix whose patterns might be hidden by this color mapping.\n\nUse `suppressMessages()` to turn off this message.") |
|
100 |
+ message_wrap("The automatically generated colors map from the 1^st and 99^th of the values in the matrix. There are outliers in the matrix whose patterns might be hidden by this color mapping. You can manually set the color to `col` argument.\n\nUse `suppressMessages()` to turn off this message.") |
|
101 | 101 |
} |
102 | 102 |
} |
103 | 103 |
} else { |
... | ... |
@@ -1006,10 +1006,33 @@ to_unit = function(str) { |
1006 | 1006 |
} |
1007 | 1007 |
|
1008 | 1008 |
|
1009 |
-resize_matrix = function(mat, nr, nc) { |
|
1009 |
+# nr <= nrow(mat) |
|
1010 |
+# nc <- ncol(mat) |
|
1011 |
+resize_matrix = function(mat, nr, nc, fun = median) { |
|
1010 | 1012 |
w_ratio = nc/ncol(mat) |
1011 | 1013 |
h_ratio = nr/nrow(mat) |
1012 |
- mat[ ceiling(1:nr / h_ratio), ceiling(1:nc / w_ratio), drop = FALSE] |
|
1014 |
+ |
|
1015 |
+ ind_r2 = ceiling(1:nr / h_ratio) |
|
1016 |
+ ind_r1 = c(1, ind_r2[-length(ind_r2)]+1) |
|
1017 |
+ ind_c2 = ceiling(1:nc / w_ratio) |
|
1018 |
+ ind_c1 = c(1, ind_c2[-length(ind_c2)]+1) |
|
1019 |
+ if(is.null(fun)) { |
|
1020 |
+ mat[ ceiling(1:nr / h_ratio), ceiling(1:nc / w_ratio), drop = FALSE] |
|
1021 |
+ } else { |
|
1022 |
+ |
|
1023 |
+ nr_reduced = length(ind_r1) |
|
1024 |
+ nc_reduced = length(ind_c1) |
|
1025 |
+ |
|
1026 |
+ ind_grid = expand.grid(1:nr_reduced, 1:nc_reduced) |
|
1027 |
+ mat_reduced = matrix(nrow = nr_reduced, ncol = nc_reduced) |
|
1028 |
+ for(k in seq_len(nrow(ind_grid))) { |
|
1029 |
+ i = ind_grid[k, 1] |
|
1030 |
+ j = ind_grid[k, 2] |
|
1031 |
+ subm = mat[seq(ind_r1[i], ind_r2[i]), seq(ind_c1[j], ind_c1[j]), drop = FALSE] |
|
1032 |
+ mat_reduced[i, j] = fun(subm) |
|
1033 |
+ } |
|
1034 |
+ return(mat_reduced) |
|
1035 |
+ } |
|
1013 | 1036 |
} |
1014 | 1037 |
|
1015 | 1038 |
|
... | ... |
@@ -1053,3 +1053,17 @@ colorRamp2_biv = function(f1, f2, transparency = 0.5) { |
1053 | 1053 |
} |
1054 | 1054 |
} |
1055 | 1055 |
|
1056 |
+ |
|
1057 |
+is_RStudio_current_dev = function() { |
|
1058 |
+ dv = names(dev.list()) |
|
1059 |
+ if(length(dv) < 2) { |
|
1060 |
+ FALSE |
|
1061 |
+ } else { |
|
1062 |
+ n = length(dv) |
|
1063 |
+ if(dv[n-1] == "RStudioGD") { |
|
1064 |
+ TRUE |
|
1065 |
+ } else { |
|
1066 |
+ FALSE |
|
1067 |
+ } |
|
1068 |
+ } |
|
1069 |
+} |
... | ... |
@@ -80,6 +80,10 @@ default_col = function(x, main_matrix = FALSE) { |
80 | 80 |
if(length(unique(x)) >= 100) { |
81 | 81 |
q1 = quantile(abs(x), 0.99) |
82 | 82 |
col_fun = colorRamp2(c(-q1, 0, q1), c("blue", "#EEEEEE", "red")) |
83 |
+ |
|
84 |
+ if(any(x > q1*3 | x < -q1*3)) { |
|
85 |
+ message_wrap("The automatically generated color maps from the minus and plus 99^th of the absolute values in the matrix. There are outliers in the matrix whose patterns might be hidden by this color mapping.\n\nUse `suppressMessages()` to turn off this message.") |
|
86 |
+ } |
|
83 | 87 |
} else { |
84 | 88 |
q1 = max(abs(x)) |
85 | 89 |
col_fun = colorRamp2(c(-q1, 0, q1), c("blue", "#EEEEEE", "red")) |
... | ... |
@@ -92,6 +96,9 @@ default_col = function(x, main_matrix = FALSE) { |
92 | 96 |
col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
93 | 97 |
} else { |
94 | 98 |
col_fun = colorRamp2(seq(q1, q2, length = 3), c("blue", "#EEEEEE", "red")) |
99 |
+ if(any(x > q2 + (q2-q1) | x < q1 - (q2-q1))) { |
|
100 |
+ message_wrap("The automatically generated color maps from the 1^st and 99^th of the values in the matrix. There are outliers in the matrix whose patterns might be hidden by this color mapping.\n\nUse `suppressMessages()` to turn off this message.") |
|
101 |
+ } |
|
95 | 102 |
} |
96 | 103 |
} else { |
97 | 104 |
col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
... | ... |
@@ -326,16 +326,22 @@ rep.list = function(x, n) { |
326 | 326 |
# == title |
327 | 327 |
# List All Heatmap Components |
328 | 328 |
# |
329 |
+# == param |
|
330 |
+# -pattern A regular expression. |
|
331 |
+# |
|
329 | 332 |
# == value |
330 | 333 |
# A vector of viewport names. |
331 | 334 |
# |
332 |
-list_components = function() { |
|
335 |
+list_components = function(pattern = NULL) { |
|
333 | 336 |
vp = grid.ls(viewports = TRUE, grobs = FALSE, flatten = FALSE, print = FALSE) |
334 | 337 |
vp = unlist(vp) |
335 | 338 |
attributes(vp) = NULL |
336 | 339 |
vp = vp[!grepl("^\\d+$", vp)] |
337 | 340 |
vp = vp[!grepl("GRID.VP", vp)] |
338 | 341 |
# unique(vp) |
342 |
+ if(!is.null(pattern)) { |
|
343 |
+ vp = grep(pattern, vp, value = TRUE) |
|
344 |
+ } |
|
339 | 345 |
vp |
340 | 346 |
} |
341 | 347 |
|
... | ... |
@@ -179,6 +179,8 @@ get_dist = function(matrix, method) { |
179 | 179 |
} else { |
180 | 180 |
stop_wrap("Since your distance method is a function, it can only accept one or two arguments.") |
181 | 181 |
} |
182 |
+ } else if(inherits(method, "dist")) { |
|
183 |
+ return(method) |
|
182 | 184 |
} else if(method %in% c("euclidean", "maximum", "manhattan", "canberra", "binary", "minkowski")) { |
183 | 185 |
# if(any(is.na(matrix))) { |
184 | 186 |
# dst = get_dist(matrix, function(x, y) { |
... | ... |
@@ -206,6 +208,8 @@ get_dist = function(matrix, method) { |
206 | 208 |
spearman = as.dist(1 - cor(t(matrix), method = "spearman")), |
207 | 209 |
kendall = as.dist(1 - cor(t(matrix), method = "kendall"))) |
208 | 210 |
} |
211 |
+ } else { |
|
212 |
+ stop_wrap(qq("method @{method} not supported")) |
|
209 | 213 |
} |
210 | 214 |
return(dst) |
211 | 215 |
} |
... | ... |
@@ -996,158 +996,6 @@ resize_matrix = function(mat, nr, nc) { |
996 | 996 |
} |
997 | 997 |
|
998 | 998 |
|
999 |
-# == title |
|
1000 |
-# Adjust positions of rectanglar shapes |
|
1001 |
-# |
|
1002 |
-# == param |
|
1003 |
-# -start position which corresponds to the start (bottom or left) of the rectangle-shapes. |
|
1004 |
-# -end position which corresponds to the end (top or right) of the rectanglar shapes. |
|
1005 |
-# -range data ranges (the minimal and maximal values) |
|
1006 |
-# -plot Whether plot the correspondance between the original positions and the adjusted positions. Only for testing. |
|
1007 |
-# |
|
1008 |
-# == details |
|
1009 |
-# This is an improved version of the `circlize::smartAlign`. |
|
1010 |
-# |
|
1011 |
-# It adjusts the positions of the rectangular shapes to make them do not overlap |
|
1012 |
-# |
|
1013 |
-# == example |
|
1014 |
-# range = c(0, 10) |
|
1015 |
-# pos1 = rbind(c(1, 2), c(5, 7)) |
|
1016 |
-# smartAlign2(pos1, range = range, plot = TRUE) |
|
1017 |
-# |
|
1018 |
-# range = c(0, 10) |
|
1019 |
-# pos1 = rbind(c(-0.5, 2), c(5, 7)) |
|
1020 |
-# smartAlign2(pos1, range = range, plot = TRUE) |
|
1021 |
-# |
|
1022 |
-# pos1 = rbind(c(-1, 2), c(3, 4), c(5, 6), c(7, 11)) |
|
1023 |
-# pos1 = pos1 + runif(length(pos1), max = 0.3, min = -0.3) |
|
1024 |
-# omfrow = par("mfrow") |
|
1025 |
-# par(mfrow = c(3, 3)) |
|
1026 |
-# for(i in 1:9) { |
|
1027 |
-# ind = sample(4, 4) |
|
1028 |
-# smartAlign2(pos1[ind, ], range = range, plot = TRUE) |
|
1029 |
-# } |
|
1030 |
-# par(mfrow = omfrow) |
|
1031 |
-# |
|
1032 |
-# pos1 = rbind(c(3, 6), c(4, 7)) |
|
1033 |
-# smartAlign2(pos1, range = range, plot = TRUE) |
|
1034 |
-# |
|
1035 |
-# pos1 = rbind(c(1, 8), c(3, 10)) |
|
1036 |
-# smartAlign2(pos1, range = range, plot = TRUE) |
|
1037 |
-# |
|
1038 |
-smartAlign2 = function(start, end, range, plot = FALSE) { |
|
1039 |
- |
|
1040 |
- if(missing(end)) { |
|
1041 |
- x1 = start[, 1] |
|
1042 |
- x2 = start[, 2] |
|
1043 |
- } else { |
|
1044 |
- x1 = start |
|
1045 |
- x2 = end |
|
1046 |
- } |
|
1047 |
- |
|
1048 |
- if(missing(range)) { |
|
1049 |
- range = range(c(x1, x2)) |
|
1050 |
- } |
|
1051 |
- |
|
1052 |
- make_plot = function(pos1, pos2, main = "") { |
|
1053 |
- oxpd = par("xpd") |
|
1054 |
- par(xpd = NA) |
|
1055 |
- plot(NULL, xlim = c(0, 4), ylim = range(c(pos1, pos2)), ann = FALSE, axes = FALSE) |
|
1056 |
- col = rand_color(nrow(pos1), transparency = 0.5) |
|
1057 |
- rect(0.5, pos1[, 1], 1.5, pos1[, 2], col = col) |
|
1058 |
- rect(2.5, pos2[, 1], 3.5, pos2[, 2], col = col) |
|
1059 |
- segments(1.5, rowMeans(pos1), 2.5, rowMeans(pos2)) |
|
1060 |
- text(1, -0.02, "original", adj = c(0.5, 1)) |
|
1061 |
- text(3, -0.02, "adjusted", adj = c(0.5, 1)) |
|
1062 |
- title(main) |
|
1063 |
- par(xpd = oxpd) |
|
1064 |
- } |
|
1065 |
- |
|
1066 |
- |
|
1067 |
- od = order(x1) |
|
1068 |
- rk = rank(x1, ties.method = "random") |
|
1069 |
- x1 = x1[od] |
|
1070 |
- x2 = x2[od] |
|
1071 |
- mid = (x1 + x2)/2 |
|
1072 |
- h = x2 - x1 |
|
1073 |
- n = length(x1) |
|
1074 |
- |
|
1075 |
- ox1 = x1 |
|
1076 |
- ox2 = x2 |
|
1077 |
- |
|
1078 |
- # sum of box heights exceeds range |
|
1079 |
- if(sum(h) > range[2] - range[1]) { |
|
1080 |
- a = ((range[2] - h[n]/2) - (range[1] + h[1]/2))/(n-1) |
|
1081 |
- m = range[1] + 1:n*a |
|
1082 |
- nx1 = m - h/2 |
|
1083 |
- nx2 = m + h/2 |
|
1084 |
- |
|
1085 |
- if(plot) { |
|
1086 |
- make_plot(cbind(ox1, ox2), cbind(nx1, nx2), main = "sum of box heights exceeds range") |
|
1087 |
- } |
|
1088 |
- |
|
1089 |
- df = data.frame(start = x1, end = x2) |
|
1090 |
- return(df[rk, , drop = FALSE]) |
|
1091 |
- } |
|
1092 |
- |
|
1093 |
- ncluster.before = -1 |
|
1094 |
- ncluster = length(x1) |
|
1095 |
- i_try = 1 |
|
1096 |
- while(ncluster.before != ncluster) { |
|
1097 |
- |
|
1098 |
- if(i_try > 100) break |
|
1099 |
- |
|
1100 |
- ncluster.before = ncluster |
|
1101 |
- cluster = rep(0, length(x1)) |
|
1102 |
- i_cluster = 1 |
|
1103 |
- cluster[1] = i_cluster |
|
1104 |
- for(i in seq_along(x1)[-1]) { |
|
1105 |
- # overlap with previous one |
|
1106 |
- if(x1[i] <= x2[i-1]) { # this means x1 should be sorted increasingly |
|
1107 |
- cluster[i] = i_cluster |
|
1108 |
- } else { |
|
1109 |
- i_cluster = i_cluster + 1 |
|
1110 |
- cluster[i] = i_cluster |
|
1111 |
- } |
|
1112 |
- } |
|
1113 |
- ncluster = length(unique(cluster)) |
|
1114 |
- |
|
1115 |
- # tile intervals in each cluster and re-assign x1 and x2 |
|
1116 |
- new_x1 = numeric(length(x1)) |
|
1117 |
- new_x2 = numeric(length(x2)) |
|
1118 |
- for(i_cluster in unique(cluster)) { |
|
1119 |
- index = which(cluster == i_cluster) |
|
1120 |
- box_height = sum(h[index]) # sum of the height in the cluster |
|
1121 |
- box_mid = (min(x1[index]) + max(x2[index]))/2 |
|
1122 |
- box_x1 = box_mid - box_height/2 |
|
1123 |
- box_x2 = box_mid + box_height/2 |
|
1124 |
- |
|
1125 |
- if(box_x1 < range[1]) { # if it exceed the bottom |
|
1126 |
- new_x2[index] = range[1] + cumsum(h[index]) |
|
1127 |
- new_x1[index] = new_x2[index] - h[index] |
|
1128 |
- } else if(box_x2 > range[2]) { |
|
1129 |
- new_x1[index] = range[2] - rev(cumsum(h[index])) |
|
1130 |
- new_x2[index] = new_x1[index] + h[index] |
|
1131 |
- } else { |
|
1132 |
- new_x2[index] = box_x1 + cumsum(h[index]) |
|
1133 |
- new_x1[index] = new_x2[index] - h[index] |
|
1134 |
- } |
|
1135 |
- } |
|
1136 |
- |
|
1137 |
- x1 = new_x1 |
|
1138 |
- x2 = new_x2 |
|
1139 |
- |
|
1140 |
- if(plot) { |
|
1141 |
- make_plot(cbind(ox1, ox2), cbind(x1, x2), main = qq("@{i_try}th try, @{ncluster} clusters")) |
|
1142 |
- } |
|
1143 |
- |
|
1144 |
- i_try = i_try + 1 |
|
1145 |
- } |
|
1146 |
- |
|
1147 |
- df = data.frame(start = x1, end = x2) |
|
1148 |
- df[rk, , drop = FALSE] |
|
1149 |
-} |
|
1150 |
- |
|
1151 | 999 |
color_overlap = function (r0, g0, b0, r, g, b, alpha = 1) { |
1152 | 1000 |
l_na_1 = is.na(r0) | is.na(g0) | is.na(b0) |
1153 | 1001 |
l_na_2 = is.na(r) | is.na(g) | is.na(b) |
... | ... |
@@ -627,6 +627,44 @@ recycle_param = function(x, all_names, default, as.list = FALSE) { |
627 | 627 |
} |
628 | 628 |
} |
629 | 629 |
|
630 |
+# recycle_list(list(a = 1), "a") |
|
631 |
+# recycle_list(1, c("a", "b")) |
|
632 |
+# recycle_list(list(a = 1), c("a", "b"), 0) |
|
633 |
+recycle_list = function(x, all_names, default = NULL) { |
|
634 |
+ n = length(all_names) |
|
635 |
+ if(is.null(x)) { |
|
636 |
+ lt = rep(list(default), n) |
|
637 |
+ names(lt) = all_names |
|
638 |
+ return(lt) |
|
639 |
+ } |
|
640 |
+ if(length(x) == 1 && !is.list(x) && n == 1) { |
|
641 |
+ lt = list(x) |
|
642 |
+ names(lt) = all_names |
|
643 |
+ return(lt) |
|
644 |
+ } |
|
645 |
+ if(length(x) == 1 && !is.list(x)) { |
|
646 |
+ lt = rep(list(x), n) |
|
647 |
+ names(lt) = all_names |
|
648 |
+ return(lt) |
|
649 |
+ } |
|
650 |
+ if(is.list(x)) { |
|
651 |
+ lt = rep(list(default), n) |
|
652 |
+ names(lt) = all_names |
|
653 |
+ for(nm in names(x)) { |
|
654 |
+ lt[[nm]] = x[[nm]] |
|
655 |
+ } |
|
656 |
+ return(lt) |
|
657 |
+ } |
|
658 |
+ if(length(x) == n) { |
|
659 |
+ lt = lapply(x, function(y) y) |
|
660 |
+ names(lt) = all_names |
|
661 |
+ return(lt) |
|
662 |
+ } |
|
663 |
+ |
|
664 |
+ stop_wrap("wrong input data type.") |
|
665 |
+ |
|
666 |
+} |
|
667 |
+ |
|
630 | 668 |
# == title |
631 | 669 |
# Convert XY in a Parent Viewport |
632 | 670 |
# |
... | ... |
@@ -965,7 +965,7 @@ resize_matrix = function(mat, nr, nc) { |
965 | 965 |
# -start position which corresponds to the start (bottom or left) of the rectangle-shapes. |
966 | 966 |
# -end position which corresponds to the end (top or right) of the rectanglar shapes. |
967 | 967 |
# -range data ranges (the minimal and maximal values) |
968 |
-# -range_fixed Whether the range is fixed for ``range`` when adjust the positions? |
|
968 |
+# -plot Whether plot the correspondance between the original positions and the adjusted positions. Only for testing. |
|
969 | 969 |
# |
970 | 970 |
# == details |
971 | 971 |
# This is an improved version of the `circlize::smartAlign`. |
... | ... |
@@ -973,25 +973,13 @@ resize_matrix = function(mat, nr, nc) { |
973 | 973 |
# It adjusts the positions of the rectangular shapes to make them do not overlap |
974 | 974 |
# |
975 | 975 |
# == example |
976 |
-# require(circlize) |
|
977 |
-# make_plot = function(pos1, pos2, range) { |
|
978 |
-# oxpd = par("xpd") |
|
979 |
-# par(xpd = NA) |
|
980 |
-# plot(NULL, xlim = c(0, 4), ylim = range, ann = FALSE) |
|
981 |
-# col = rand_color(nrow(pos1), transparency = 0.5) |
|
982 |
-# rect(0.5, pos1[, 1], 1.5, pos1[, 2], col = col) |
|
983 |
-# rect(2.5, pos2[, 1], 3.5, pos2[, 2], col = col) |
|
984 |
-# segments(1.5, rowMeans(pos1), 2.5, rowMeans(pos2)) |
|
985 |
-# par(xpd = oxpd) |
|
986 |
-# } |
|
987 |
-# |
|
988 | 976 |
# range = c(0, 10) |
989 | 977 |
# pos1 = rbind(c(1, 2), c(5, 7)) |
990 |
-# make_plot(pos1, smartAlign2(pos1, range = range), range) |
|
978 |
+# smartAlign2(pos1, range = range, plot = TRUE) |
|
991 | 979 |
# |
992 | 980 |
# range = c(0, 10) |
993 | 981 |
# pos1 = rbind(c(-0.5, 2), c(5, 7)) |
994 |
-# make_plot(pos1, smartAlign2(pos1, range = range), range) |
|
982 |
+# smartAlign2(pos1, range = range, plot = TRUE) |
|
995 | 983 |
# |
996 | 984 |
# pos1 = rbind(c(-1, 2), c(3, 4), c(5, 6), c(7, 11)) |
997 | 985 |
# pos1 = pos1 + runif(length(pos1), max = 0.3, min = -0.3) |
... | ... |
@@ -999,18 +987,17 @@ resize_matrix = function(mat, nr, nc) { |
999 | 987 |
# par(mfrow = c(3, 3)) |
1000 | 988 |
# for(i in 1:9) { |
1001 | 989 |
# ind = sample(4, 4) |
1002 |
-# make_plot(pos1[ind, ], smartAlign2(pos1[ind, ], range = range), range) |
|
990 |
+# smartAlign2(pos1[ind, ], range = range, plot = TRUE) |
|
1003 | 991 |
# } |
1004 | 992 |
# par(mfrow = omfrow) |
1005 | 993 |
# |
1006 | 994 |
# pos1 = rbind(c(3, 6), c(4, 7)) |
1007 |
-# make_plot(pos1, smartAlign2(pos1, range = range), range) |
|
995 |
+# smartAlign2(pos1, range = range, plot = TRUE) |
|
1008 | 996 |
# |
1009 | 997 |
# pos1 = rbind(c(1, 8), c(3, 10)) |
1010 |
-# make_plot(pos1, smartAlign2(pos1, range = range), range) |
|
1011 |
-# make_plot(pos1, smartAlign2(pos1, range = range, range_fixed = FALSE), range) |
|
998 |
+# smartAlign2(pos1, range = range, plot = TRUE) |
|
1012 | 999 |
# |
1013 |
-smartAlign2 = function(start, end, range, range_fixed = TRUE) { |
|
1000 |
+smartAlign2 = function(start, end, range, plot = FALSE) { |
|
1014 | 1001 |
|
1015 | 1002 |
if(missing(end)) { |
1016 | 1003 |
x1 = start[, 1] |
... | ... |
@@ -1024,16 +1011,54 @@ smartAlign2 = function(start, end, range, range_fixed = TRUE) { |
1024 | 1011 |
range = range(c(x1, x2)) |
1025 | 1012 |
} |
1026 | 1013 |
|
1014 |
+ make_plot = function(pos1, pos2, main = "") { |
|
1015 |
+ oxpd = par("xpd") |
|
1016 |
+ par(xpd = NA) |
|
1017 |
+ plot(NULL, xlim = c(0, 4), ylim = range(c(pos1, pos2)), ann = FALSE, axes = FALSE) |
|
1018 |
+ col = rand_color(nrow(pos1), transparency = 0.5) |
|
1019 |
+ rect(0.5, pos1[, 1], 1.5, pos1[, 2], col = col) |
|
1020 |
+ rect(2.5, pos2[, 1], 3.5, pos2[, 2], col = col) |
|
1021 |
+ segments(1.5, rowMeans(pos1), 2.5, rowMeans(pos2)) |
|
1022 |
+ text(1, -0.02, "original", adj = c(0.5, 1)) |
|
1023 |
+ text(3, -0.02, "adjusted", adj = c(0.5, 1)) |
|
1024 |
+ title(main) |
|
1025 |
+ par(xpd = oxpd) |
|
1026 |
+ } |
|
1027 |
+ |
|
1028 |
+ |
|
1027 | 1029 |
od = order(x1) |
1028 | 1030 |
rk = rank(x1, ties.method = "random") |
1029 | 1031 |
x1 = x1[od] |
1030 | 1032 |
x2 = x2[od] |
1033 |
+ mid = (x1 + x2)/2 |
|
1031 | 1034 |
h = x2 - x1 |
1035 |
+ n = length(x1) |
|
1036 |
+ |
|
1037 |
+ ox1 = x1 |
|
1038 |
+ ox2 = x2 |
|
1032 | 1039 |
|
1040 |
+ # sum of box heights exceeds range |
|
1041 |
+ if(sum(h) > range[2] - range[1]) { |
|
1042 |
+ a = ((range[2] - h[n]/2) - (range[1] + h[1]/2))/(n-1) |
|
1043 |
+ m = range[1] + 1:n*a |
|
1044 |
+ nx1 = m - h/2 |
|
1045 |
+ nx2 = m + h/2 |
|
1046 |
+ |
|
1047 |
+ if(plot) { |
|
1048 |
+ make_plot(cbind(ox1, ox2), cbind(nx1, nx2), main = "sum of box heights exceeds range") |
|
1049 |
+ } |
|
1050 |
+ |
|
1051 |
+ df = data.frame(start = x1, end = x2) |
|
1052 |
+ return(df[rk, , drop = FALSE]) |
|
1053 |
+ } |
|
1054 |
+ |
|
1033 | 1055 |
ncluster.before = -1 |
1034 | 1056 |
ncluster = length(x1) |
1035 |
- i_try = 0 |
|
1057 |
+ i_try = 1 |
|
1036 | 1058 |
while(ncluster.before != ncluster) { |
1059 |
+ |
|
1060 |
+ if(i_try > 100) break |
|
1061 |
+ |
|
1037 | 1062 |
ncluster.before = ncluster |
1038 | 1063 |
cluster = rep(0, length(x1)) |
1039 | 1064 |
i_cluster = 1 |
... | ... |
@@ -1048,43 +1073,35 @@ smartAlign2 = function(start, end, range, range_fixed = TRUE) { |
1048 | 1073 |
} |
1049 | 1074 |
} |
1050 | 1075 |
ncluster = length(unique(cluster)) |
1051 |
- |
|
1052 |
- if(ncluster.before == ncluster && i_try > 0) break |
|
1053 | 1076 |
|
1054 |
- if(i_try > 100) break |
|
1055 |
- |
|
1056 | 1077 |
# tile intervals in each cluster and re-assign x1 and x2 |
1057 | 1078 |
new_x1 = numeric(length(x1)) |
1058 | 1079 |
new_x2 = numeric(length(x2)) |
1059 | 1080 |
for(i_cluster in unique(cluster)) { |
1060 | 1081 |
index = which(cluster == i_cluster) |
1061 |
- total_len = sum(x2[index] - x1[index]) # sum of the height in the cluster |
|
1062 |
- mid = (min(x1[index]) + max(x2[index]))/2 |
|
1063 |
- if(total_len > range[2] - range[1]) { |
|
1064 |
- # tp = seq(range[1], range[2], length = length(index) + 1) |
|
1065 |
- if(range_fixed) { |
|
1066 |
- tp = cumsum(c(0, h[index]/sum(h[index])))*(range[2] - range[1]) + range[1] |
|
1067 |
- } else { |
|
1068 |
- tp = c(0, cumsum(h[index])) + mid - sum(h[index])/2 |
|
1069 |
- } |
|
1070 |
- } else if(mid - total_len/2 < range[1]) { # if it exceed the bottom |
|
1071 |
- # tp = seq(range[1], range[1] + total_len, length = length(index) + 1) |
|
1072 |
- tp = c(0, cumsum(h[index])) + range[1] |
|
1073 |
- } else if(mid + total_len/2 > range[2]) { |
|
1074 |
- # tp = seq(range[2] - total_len, range[2], length = length(index) + 1) |
|
1075 |
- tp = range[2] - rev(c(0, cumsum(h[index]))) |
|
1082 |
+ box_height = sum(h[index]) # sum of the height in the cluster |
|
1083 |
+ box_mid = (min(x1[index]) + max(x2[index]))/2 |
|
1084 |
+ box_x1 = box_mid - box_height/2 |
|
1085 |
+ box_x2 = box_mid + box_height/2 |
|
1086 |
+ |
|
1087 |
+ if(box_x1 < range[1]) { # if it exceed the bottom |
|
1088 |
+ new_x2[index] = range[1] + cumsum(h[index]) |
|
1089 |
+ new_x1[index] = new_x2[index] - h[index] |
|
1090 |
+ } else if(box_x2 > range[2]) { |
|
1091 |
+ new_x1[index] = range[2] - rev(cumsum(h[index])) |
|
1092 |
+ new_x2[index] = new_x1[index] + h[index] |
|
1076 | 1093 |
} else { |
1077 |
- # tp = seq(mid - total_len/2, mid + total_len/2, length = length(index)+1) |
|
1078 |
- tp = c(0, cumsum(h[index])) + mid - sum(h[index])/2 |
|
1094 |
+ new_x2[index] = box_x1 + cumsum(h[index]) |
|
1095 |
+ new_x1[index] = new_x2[index] - h[index] |
|
1079 | 1096 |
} |
1080 |
- new_x1[index] = tp[-length(tp)] |
|
1081 |
- new_x2[index] = tp[-1] |
|
1082 | 1097 |
} |
1083 |
- mid = (new_x1 + new_x2)/2 |
|
1084 |
- h = (x2 - x1) |
|
1085 |
- |
|
1086 |
- x1 = mid - h/2 |
|
1087 |
- x2 = mid + h/2 |
|
1098 |
+ |
|
1099 |
+ x1 = new_x1 |
|
1100 |
+ x2 = new_x2 |
|
1101 |
+ |
|
1102 |
+ if(plot) { |
|
1103 |
+ make_plot(cbind(ox1, ox2), cbind(x1, x2), main = qq("@{i_try}th try, @{ncluster} clusters")) |
|
1104 |
+ } |
|
1088 | 1105 |
|
1089 | 1106 |
i_try = i_try + 1 |
1090 | 1107 |
} |
... | ... |
@@ -545,6 +545,10 @@ unit_in_mm = function(x) { |
545 | 545 |
identical(unitType(x), "mm") |
546 | 546 |
} |
547 | 547 |
|
548 |
+unit_to_numeric = function(x) { |
|
549 |
+ as.numeric(x) |
|
550 |
+} |
|
551 |
+ |
|
548 | 552 |
normalize_graphic_param_to_mat = function(x, nc, nr, name) { |
549 | 553 |
if(is.matrix(x)) { |
550 | 554 |
if(nrow(x) == nr && ncol(x) == nc) { |
... | ... |
@@ -528,7 +528,7 @@ unit.c = function(...) { |
528 | 528 |
if(!unit_in_mm(y)) { |
529 | 529 |
stop_wrap("y should be in mm unit") |
530 | 530 |
} |
531 |
- x[[1]] > y[[1]] |
|
531 |
+ as.numeric(x) > as.numeric(y) |
|
532 | 532 |
} |
533 | 533 |
|
534 | 534 |
"<.unit" = function(x, y) { |
... | ... |
@@ -538,16 +538,11 @@ unit.c = function(...) { |
538 | 538 |
if(!unit_in_mm(y)) { |
539 | 539 |
stop_wrap("y should be in mm unit") |
540 | 540 |
} |
541 |
- x[[1]] < y[[1]] |
|
541 |
+ as.numeric(x) < as.numeric(y) |
|
542 | 542 |
} |
543 | 543 |
|
544 | 544 |
unit_in_mm = function(x) { |
545 |
- if(getRversion() >= "4.0.0") { |
|
546 |
- unitType = get("unitType", envir = asNamespace("grid")) |
|
547 |
- identical(unitType(x), "mm") |
|
548 |
- } else { |
|
549 |
- identical(attr(x, "unit"), "mm") |
|
550 |
- } |
|
545 |
+ identical(unitType(x), "mm") |
|
551 | 546 |
} |
552 | 547 |
|
553 | 548 |
normalize_graphic_param_to_mat = function(x, nc, nr, name) { |
... | ... |
@@ -522,25 +522,34 @@ unit.c = function(...) { |
522 | 522 |
} |
523 | 523 |
|
524 | 524 |
">.unit" = function(x, y) { |
525 |
- if(!identical(attr(x, "unit"), "mm")) { |
|
525 |
+ if(!unit_in_mm(x)) { |
|
526 | 526 |
stop_wrap("x should be in mm unit") |
527 | 527 |
} |
528 |
- if(!identical(attr(y, "unit"), "mm")) { |
|
528 |
+ if(!unit_in_mm(y)) { |
|
529 | 529 |
stop_wrap("y should be in mm unit") |
530 | 530 |
} |
531 | 531 |
x[[1]] > y[[1]] |
532 | 532 |
} |
533 | 533 |
|
534 | 534 |
"<.unit" = function(x, y) { |
535 |
- if(!identical(attr(x, "unit"), "mm")) { |
|
535 |
+ if(!unit_in_mm(x)) { |
|
536 | 536 |
stop_wrap("x should be in mm unit") |
537 | 537 |
} |
538 |
- if(!identical(attr(y, "unit"), "mm")) { |
|
538 |
+ if(!unit_in_mm(y)) { |
|
539 | 539 |
stop_wrap("y should be in mm unit") |
540 | 540 |
} |
541 | 541 |
x[[1]] < y[[1]] |
542 | 542 |
} |
543 | 543 |
|
544 |
+unit_in_mm = function(x) { |
|
545 |
+ if(getRversion() >= "4.0.0") { |
|
546 |
+ unitType = get("unitType", envir = asNamespace("grid")) |
|
547 |
+ identical(unitType(x), "mm") |
|
548 |
+ } else { |
|
549 |
+ identical(attr(x, "unit"), "mm") |
|
550 |
+ } |
|
551 |
+} |
|
552 |
+ |
|
544 | 553 |
normalize_graphic_param_to_mat = function(x, nc, nr, name) { |
545 | 554 |
if(is.matrix(x)) { |
546 | 555 |
if(nrow(x) == nr && ncol(x) == nc) { |
... | ... |
@@ -420,20 +420,6 @@ text_width = function(text, gp = gpar()) { |
420 | 420 |
convertWidth(u, "mm") |
421 | 421 |
} |
422 | 422 |
|
423 |
-grid.text = function(label, x = unit(0.5, "npc"), y = unit(0.5, "npc"), |
|
424 |
- just = "centre", hjust = NULL, vjust = NULL, rot = 0, check.overlap = FALSE, |
|
425 |
- default.units = "npc", name = NULL, gp = gpar(), draw = TRUE, |
|
426 |
- vp = NULL) { |
|
427 |
- tg <- textGrob(label = label, x = x, y = y, just = just, |
|
428 |
- hjust = hjust, vjust = vjust, rot = rot, check.overlap = check.overlap, |
|
429 |
- default.units = default.units, name = name, gp = gp, |
|
430 |
- vp = vp) |
|
431 |
- tw = text_width(label) |
|
432 |
- th = text_height(label) |
|
433 |
- grid.draw(tg) |
|
434 |
- if(identical(just, "")) |
|
435 |
-} |
|
436 |
- |
|
437 | 423 |
text_height = function(text, gp = gpar()) { |
438 | 424 |
if(is.null(text)) { |
439 | 425 |
return(unit(0, "mm")) |
... | ... |
@@ -409,6 +409,42 @@ max_text_height = function(text, gp = gpar(), rot = 0) { |
409 | 409 |
convertHeight(u, "mm") |
410 | 410 |
} |
411 | 411 |
|
412 |
+text_width = function(text, gp = gpar()) { |
|
413 |
+ if(is.null(text)) { |
|
414 |
+ return(unit(0, "mm")) |
|
415 |
+ } |
|
416 |
+ n = length(text) |
|
417 |
+ gp = recycle_gp(gp, n) |
|
418 |
+ |
|
419 |
+ u = do.call("unit.c", lapply(seq_len(n), function(i) grobWidth(textGrob(text[i], gp = subset_gp(gp, i))))) |
|
420 |
+ convertWidth(u, "mm") |
|
421 |
+} |
|
422 |
+ |
|
423 |
+grid.text = function(label, x = unit(0.5, "npc"), y = unit(0.5, "npc"), |
|
424 |
+ just = "centre", hjust = NULL, vjust = NULL, rot = 0, check.overlap = FALSE, |
|
425 |
+ default.units = "npc", name = NULL, gp = gpar(), draw = TRUE, |
|
426 |
+ vp = NULL) { |
|
427 |
+ tg <- textGrob(label = label, x = x, y = y, just = just, |
|
428 |
+ hjust = hjust, vjust = vjust, rot = rot, check.overlap = check.overlap, |
|
429 |
+ default.units = default.units, name = name, gp = gp, |
|
430 |
+ vp = vp) |
|
431 |
+ tw = text_width(label) |
|
432 |
+ th = text_height(label) |
|
433 |
+ grid.draw(tg) |
|
434 |
+ if(identical(just, "")) |
|
435 |
+} |
|
436 |
+ |
|
437 |
+text_height = function(text, gp = gpar()) { |
|
438 |
+ if(is.null(text)) { |
|
439 |
+ return(unit(0, "mm")) |
|
440 |
+ } |
|
441 |
+ n = length(text) |
|
442 |
+ gp = recycle_gp(gp, n) |
|
443 |
+ |
|
444 |
+ u = do.call("unit.c", lapply(seq_len(n), function(i) grobHeight(textGrob(text[i], gp = subset_gp(gp, i))))) |
|
445 |
+ convertHeight(u, "mm") |
|
446 |
+} |
|
447 |
+ |
|
412 | 448 |
dev.null = function(...) { |
413 | 449 |
pdf(file = NULL, ...) |
414 | 450 |
} |
... | ... |
@@ -341,6 +341,7 @@ list_components = function() { |
341 | 341 |
# == param |
342 | 342 |
# -text A vector of text. |
343 | 343 |
# -gp Graphic parameters for text. |
344 |
+# -rot Rotation of the text, scalar. |
|
344 | 345 |
# |
345 | 346 |
# == details |
346 | 347 |
# It simply calculates maximum width of a list of `grid::textGrob` objects. |
... | ... |
@@ -360,14 +361,14 @@ list_components = function() { |
360 | 361 |
# x = c("a", "bb", "ccc") |
361 | 362 |
# max_text_width(x, gp = gpar(fontsize = 10)) |
362 | 363 |
# |
363 |
-max_text_width = function(text, gp = gpar()) { |
|
364 |
+max_text_width = function(text, gp = gpar(), rot = 0) { |
|
364 | 365 |
if(is.null(text)) { |
365 | 366 |
return(unit(0, "mm")) |
366 | 367 |
} |
367 | 368 |
n = length(text) |
368 | 369 |
gp = recycle_gp(gp, n) |
369 | 370 |
|
370 |
- u = max(do.call("unit.c", lapply(seq_len(n), function(i) grobWidth(textGrob(text[i], gp = subset_gp(gp, i)))))) |
|
371 |
+ u = max(do.call("unit.c", lapply(seq_len(n), function(i) grobWidth(textGrob(text[i], gp = subset_gp(gp, i), rot = rot))))) |
|
371 | 372 |
convertWidth(u, "mm") |
372 | 373 |
} |
373 | 374 |
|
... | ... |
@@ -377,6 +378,7 @@ max_text_width = function(text, gp = gpar()) { |
377 | 378 |
# == param |
378 | 379 |
# -text A vector of text. |
379 | 380 |
# -gp Graphic parameters for text. |
381 |
+# -rot Rotation of the text, scalar. |
|
380 | 382 |
# |
381 | 383 |
# == details |
382 | 384 |
# It simply calculates maximum height of a list of `grid::textGrob` objects. |
... | ... |
@@ -396,14 +398,14 @@ max_text_width = function(text, gp = gpar()) { |
396 | 398 |
# x = c("a", "b\nb", "c\nc\nc") |
397 | 399 |
# max_text_height(x, gp = gpar(fontsize = 10)) |
398 | 400 |
# |
399 |
-max_text_height = function(text, gp = gpar()) { |
|
401 |
+max_text_height = function(text, gp = gpar(), rot = 0) { |
|
400 | 402 |
if(is.null(text)) { |
401 | 403 |
return(unit(0, "mm")) |
402 | 404 |
} |
403 | 405 |
n = length(text) |
404 | 406 |
gp = recycle_gp(gp, n) |
405 | 407 |
|
406 |
- u = max(do.call("unit.c", lapply(seq_len(n), function(i) grobHeight(textGrob(text[i], gp = subset_gp(gp, i)))))) |
|
408 |
+ u = max(do.call("unit.c", lapply(seq_len(n), function(i) grobHeight(textGrob(text[i], gp = subset_gp(gp, i), rot = rot))))) |
|
407 | 409 |
convertHeight(u, "mm") |
408 | 410 |
} |
409 | 411 |
|
... | ... |
@@ -1061,3 +1061,43 @@ smartAlign2 = function(start, end, range, range_fixed = TRUE) { |
1061 | 1061 |
df[rk, , drop = FALSE] |
1062 | 1062 |
} |
1063 | 1063 |
|
1064 |
+color_overlap = function (r0, g0, b0, r, g, b, alpha = 1) { |
|
1065 |
+ l_na_1 = is.na(r0) | is.na(g0) | is.na(b0) |
|
1066 |
+ l_na_2 = is.na(r) | is.na(g) | is.na(b) |
|
1067 |
+ r = ifelse(l_na_1 & l_na_2, 1, ifelse(l_na_1, r * alpha, |
|
1068 |
+ ifelse(l_na_2, r0, r * alpha + r0 * (1 - alpha)))) |
|
1069 |
+ g = ifelse(l_na_1 & l_na_2, 1, ifelse(l_na_1, g * alpha, |
|
1070 |
+ ifelse(l_na_2, g0, g * alpha + g0 * (1 - alpha)))) |
|
1071 |
+ b = ifelse(l_na_1 & l_na_2, 1, ifelse(l_na_1, b * alpha, |
|
1072 |
+ ifelse(l_na_2, b0, b * alpha + b0 * (1 - alpha)))) |
|
1073 |
+ return(list(r = r, g = g, b = b)) |
|
1074 |
+} |
|
1075 |
+ |
|
1076 |
+colorRamp2_biv = function(f1, f2, transparency = 0.5) { |
|
1077 |
+ f1 = f1 |
|
1078 |
+ f2 = f2 |
|
1079 |
+ if(length(transparency) == 1) transparency = rep(transparency, 2) |
|
1080 |
+ f = function(x1, x2) { |
|
1081 |
+ if(missing(x2)) { |
|
1082 |
+ if(ncol(x1) == 2) { |
|
1083 |
+ x2 = x1[, 2] |
|
1084 |
+ x1 = x1[, 1] |
|
1085 |
+ } else { |
|
1086 |
+ stop_wrap("If only one variable is specified, it should be a matrix/data frame with two columns.") |
|
1087 |
+ } |
|
1088 |
+ } |
|
1089 |
+ col1 = col2rgb(f1(x1), alpha = TRUE)/255 |
|
1090 |
+ col2 = col2rgb(f2(x2), alpha = TRUE)/255 |
|
1091 |
+ |
|
1092 |
+ if(length(transparency)) { |
|
1093 |
+ col1[4, ] = 1 - transparency[1] |
|
1094 |
+ col2[4, ] = 1 - transparency[2] |
|
1095 |
+ } |
|
1096 |
+ |
|
1097 |
+ col1 = col1[1:3, , drop = FALSE] * rep(col1[4, ], each = 3) |
|
1098 |
+ lt = color_overlap(col1[1, ], col1[2, ], col1[3, ], |
|
1099 |
+ col2[1, ], col2[2, ], col2[3, ], alpha = col2[4, ]) |
|
1100 |
+ rgb(lt[[1]], lt[[2]], lt[[3]]) |
|
1101 |
+ } |
|
1102 |
+} |
|
1103 |
+ |
... | ... |
@@ -73,7 +73,7 @@ default_col = function(x, main_matrix = FALSE) { |
73 | 73 |
} else if(is.numeric(x)) { |
74 | 74 |
if(main_matrix) { |
75 | 75 |
p = sum(x > 0)/sum(x != 0) |
76 |
- if(p > 0.3 & p < 0.7) { |
|
76 |
+ if(p > 0.25 & p < 0.75) { |
|
77 | 77 |
if(ht_opt$verbose) { |
78 | 78 |
cat("This matrix has both negative and positive values, use a color mapping symmetric to zero\n") |
79 | 79 |
} |
... | ... |
@@ -721,6 +721,90 @@ pindex = function(m, i, j) { |
721 | 721 |
} |
722 | 722 |
} |
723 | 723 |
|
724 |
+# == title |
|
725 |
+# Restore the index vector to index matrix in layer_fun |
|
726 |
+# |
|
727 |
+# == param |
|
728 |
+# -j Column indices directly from ``layer_fun``. |
|
729 |
+# -i Row indices directly from ``layer_fun``. |
|
730 |
+# -x Position on x-direction directly from ``layer_fun``. |
|
731 |
+# -y Position on y-direction directly from ``layer_fun``. |
|
732 |
+# |
|
733 |
+# == details |
|
734 |
+# The values that are sent to ``layer_fun`` are all vectors (for the vectorization |
|
735 |
+# of the grid graphic functions), however, the heatmap slice where |
|
736 |
+# ``layer_fun`` is applied to, is still represented by a matrix, thus, it would be |
|
737 |
+# very convinient if all the arguments in ``layer_fun`` can be converted to the |
|
738 |
+# sub-matrix for the current slice. Here, as shown in above example, |
|
739 |
+# `restore_matrix` does the job. `restore_matrix` directly accepts the first |
|
740 |
+# four argument in ``layer_fun`` and returns an index matrix, where rows and |
|
741 |
+# columns correspond to the rows and columns in the current slice, from top to |
|
742 |
+# bottom and from left to right. The values in the matrix are the natural order |
|
743 |
+# of e.g. vector ``j`` in current slice. |
|
744 |
+# |
|
745 |
+# For following code: |
|
746 |
+# |
|
747 |
+# Heatmap(small_mat, name = "mat", col = col_fun, |
|
748 |
+# row_km = 2, column_km = 2, |
|
749 |
+# layer_fun = function(j, i, x, y, w, h, fill) { |
|
750 |
+# ind_mat = restore_matrix(j, i, x, y) |
|
751 |
+# print(ind_mat) |
|
752 |
+# } |
|
753 |
+# ) |
|
754 |
+# |
|
755 |
+# The first output which is for the top-left slice: |
|
756 |
+# |
|
757 |
+# [,1] [,2] [,3] [,4] [,5] |
|
758 |
+# [1,] 1 4 7 10 13 |
|
759 |
+# [2,] 2 5 8 11 14 |
|
760 |
+# [3,] 3 6 9 12 15 |
|
761 |
+# |
|
762 |
+# As you see, this is a three-row and five-column index matrix where the first |
|
763 |
+# row corresponds to the top row in the slice. The values in the matrix |
|
764 |
+# correspond to the natural index (i.e. 1, 2, ...) in ``j``, ``i``, ``x``, ``y``, |
|
765 |
+# ... in ``layer_fun``. Now, if we want to add values on the second column in the |
|
766 |
+# top-left slice, the code which is put inside ``layer_fun`` would look like: |
|
767 |
+# |
|
768 |
+# for(ind in ind_mat[, 2]) { |
|
769 |
+# grid.text(small_mat[i[ind], j[ind]], x[ind], y[ind], ...) |
|
770 |
+# } |
|
771 |
+# |
|
772 |
+# == example |
|
773 |
+# set.seed(123) |
|
774 |
+# mat = matrix(rnorm(81), nr = 9) |
|
775 |
+# Heatmap(mat, row_km = 2, column_km = 2, |
|
776 |
+# layer_fun = function(j, i, x, y, width, height, fill) { |
|
777 |
+# ind_mat = restore_matrix(j, i, x, y) |
|
778 |
+# print(ind_mat) |
|
779 |
+# }) |
|
780 |
+# |
|
781 |
+# set.seed(123) |
|
782 |
+# mat = matrix(round(rnorm(81), 2), nr = 9) |
|
783 |
+# Heatmap(mat, row_km = 2, column_km = 2, |
|
784 |
+# layer_fun = function(j, i, x, y, width, height, fill) { |
|
785 |
+# ind_mat = restore_matrix(j, i, x, y) |
|
786 |
+# ind = unique(c(ind_mat[2, ], ind_mat[, 3])) |
|
787 |
+# grid.text(pindex(mat, i[ind], j[ind]), x[ind], y[ind]) |
|
788 |
+# }) |
|
789 |
+restore_matrix = function(j, i, x, y) { |
|
790 |
+ x = as.numeric(x) |
|
791 |
+ y = as.numeric(y) |
|
792 |
+ od = order(x, rev(y)) |
|
793 |
+ ind = seq_along(i) |
|
794 |
+ j = j[od] |
|
795 |
+ i = i[od] |
|
796 |
+ x = x[od] |
|
797 |
+ y = y[od] |
|
798 |
+ ind = ind[od] |
|
799 |
+ |
|
800 |
+ nr = length(unique(i)) |
|
801 |
+ nc = length(unique(j)) |
|
802 |
+ # I = matrix(i, nrow = nr, ncol = nc) |
|
803 |
+ # J = matrix(j, nrow = nr, ncol = nc) |
|
804 |
+ IND = matrix(ind, nrow = nr, ncol = nc) |
|
805 |
+ return(IND) |
|
806 |
+} |
|
807 |
+ |
|
724 | 808 |
|
725 | 809 |
unit_with_vp = function(..., vp = current.viewport()$name) { |
726 | 810 |
u = unit(...) |
... | ... |
@@ -72,7 +72,7 @@ default_col = function(x, main_matrix = FALSE) { |
72 | 72 |
return(colors) |
73 | 73 |
} else if(is.numeric(x)) { |
74 | 74 |
if(main_matrix) { |
75 |
- p = sum(x > 0)/length(x) |
|
75 |
+ p = sum(x > 0)/sum(x != 0) |
|
76 | 76 |
if(p > 0.3 & p < 0.7) { |
77 | 77 |
if(ht_opt$verbose) { |
78 | 78 |
cat("This matrix has both negative and positive values, use a color mapping symmetric to zero\n") |
... | ... |
@@ -72,16 +72,30 @@ default_col = function(x, main_matrix = FALSE) { |
72 | 72 |
return(colors) |
73 | 73 |
} else if(is.numeric(x)) { |
74 | 74 |
if(main_matrix) { |
75 |
- if(length(unique(x)) > 100) { |
|
76 |
- q1 = quantile(x, 0.01) |
|
77 |
- q2 = quantile(x, 0.99) |
|
78 |
- if(length(unique(x[x > q1 & x < q2])) == 1) { |
|
79 |
- col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
75 |
+ p = sum(x > 0)/length(x) |
|
76 |
+ if(p > 0.3 & p < 0.7) { |
|
77 |
+ if(ht_opt$verbose) { |
|
78 |
+ cat("This matrix has both negative and positive values, use a color mapping symmetric to zero\n") |
|
79 |
+ } |
|
80 |
+ if(length(unique(x)) >= 100) { |
|
81 |
+ q1 = quantile(abs(x), 0.99) |
|
82 |
+ col_fun = colorRamp2(c(-q1, 0, q1), c("blue", "#EEEEEE", "red")) |
|
80 | 83 |
} else { |
81 |
- col_fun = colorRamp2(seq(q1, q2, length = 3), c("blue", "#EEEEEE", "red")) |
|
84 |
+ q1 = max(abs(x)) |
|
85 |
+ col_fun = colorRamp2(c(-q1, 0, q1), c("blue", "#EEEEEE", "red")) |
|
82 | 86 |
} |
83 | 87 |
} else { |
84 |
- col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
88 |
+ if(length(unique(x)) >= 100) { |
|
89 |
+ q1 = quantile(x, 0.01) |
|
90 |
+ q2 = quantile(x, 0.99) |
|
91 |
+ if(length(unique(x[x > q1 & x < q2])) == 1) { |
|
92 |
+ col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
93 |
+ } else { |
|
94 |
+ col_fun = colorRamp2(seq(q1, q2, length = 3), c("blue", "#EEEEEE", "red")) |
|
95 |
+ } |
|
96 |
+ } else { |
|
97 |
+ col_fun = colorRamp2(seq(min(x), max(x), length = 3), c("blue", "#EEEEEE", "red")) |
|
98 |
+ } |
|
85 | 99 |
} |
86 | 100 |
} else { |
87 | 101 |
#col_fun = colorRamp2(range(min(x), max(x)), c("white", hsv(runif(1), 1, 1))) |
... | ... |
@@ -804,8 +804,8 @@ grid.boxplot = function(value, pos, outline = TRUE, box_width = 0.6, |
804 | 804 |
} |
805 | 805 |
} |
806 | 806 |
|
807 |
-random_str = function() { |
|
808 |
- paste(sample(c(letters, LETTERS, 0:9), 8), collapse = "") |
|
807 |
+random_str = function(k = 1, len = 10) { |
|
808 |
+ sapply(seq_len(k), function(i) paste(sample(c(letters, LETTERS, 0:9), len), collapse = "")) |
|
809 | 809 |
} |
810 | 810 |
|
811 | 811 |
|
... | ... |
@@ -920,6 +920,8 @@ smartAlign2 = function(start, end, range, range_fixed = TRUE) { |
920 | 920 |
ncluster = length(unique(cluster)) |
921 | 921 |
|
922 | 922 |
if(ncluster.before == ncluster && i_try > 0) break |
923 |
+ |
|
924 |
+ if(i_try > 100) break |
|
923 | 925 |
|
924 | 926 |
# tile intervals in each cluster and re-assign x1 and x2 |
925 | 927 |
new_x1 = numeric(length(x1)) |
... | ... |
@@ -826,3 +826,138 @@ resize_matrix = function(mat, nr, nc) { |
826 | 826 |
h_ratio = nr/nrow(mat) |
827 | 827 |
mat[ ceiling(1:nr / h_ratio), ceiling(1:nc / w_ratio), drop = FALSE] |
828 | 828 |
} |
829 |
+ |
|
830 |
+ |
|
831 |
+# == title |
|
832 |
+# Adjust positions of rectanglar shapes |
|
833 |
+# |
|
834 |
+# == param |
|
835 |
+# -start position which corresponds to the start (bottom or left) of the rectangle-shapes. |
|
836 |
+# -end position which corresponds to the end (top or right) of the rectanglar shapes. |
|
837 |
+# -range data ranges (the minimal and maximal values) |
|
838 |
+# -range_fixed Whether the range is fixed for ``range`` when adjust the positions? |
|
839 |
+# |
|
840 |
+# == details |
|
841 |
+# This is an improved version of the `circlize::smartAlign`. |
|
842 |
+# |
|
843 |
+# It adjusts the positions of the rectangular shapes to make them do not overlap |
|
844 |
+# |
|
845 |
+# == example |
|
846 |
+# require(circlize) |
|
847 |
+# make_plot = function(pos1, pos2, range) { |
|
848 |
+# oxpd = par("xpd") |
|
849 |
+# par(xpd = NA) |
|
850 |
+# plot(NULL, xlim = c(0, 4), ylim = range, ann = FALSE) |
|
851 |
+# col = rand_color(nrow(pos1), transparency = 0.5) |
|
852 |
+# rect(0.5, pos1[, 1], 1.5, pos1[, 2], col = col) |
|
853 |
+# rect(2.5, pos2[, 1], 3.5, pos2[, 2], col = col) |
|
854 |
+# segments(1.5, rowMeans(pos1), 2.5, rowMeans(pos2)) |
|
855 |
+# par(xpd = oxpd) |
|
856 |
+# } |
|
857 |
+# |
|
858 |
+# range = c(0, 10) |
|
859 |
+# pos1 = rbind(c(1, 2), c(5, 7)) |
|
860 |
+# make_plot(pos1, smartAlign2(pos1, range = range), range) |
|
861 |
+# |
|
862 |
+# range = c(0, 10) |
|
863 |
+# pos1 = rbind(c(-0.5, 2), c(5, 7)) |
|
864 |
+# make_plot(pos1, smartAlign2(pos1, range = range), range) |
|
865 |
+# |
|
866 |
+# pos1 = rbind(c(-1, 2), c(3, 4), c(5, 6), c(7, 11)) |
|
867 |
+# pos1 = pos1 + runif(length(pos1), max = 0.3, min = -0.3) |
|
868 |
+# omfrow = par("mfrow") |
|
869 |
+# par(mfrow = c(3, 3)) |
|
870 |
+# for(i in 1:9) { |
|
871 |
+# ind = sample(4, 4) |
|
872 |
+# make_plot(pos1[ind, ], smartAlign2(pos1[ind, ], range = range), range) |
|
873 |
+# } |
|
874 |
+# par(mfrow = omfrow) |
|
875 |
+# |
|
876 |
+# pos1 = rbind(c(3, 6), c(4, 7)) |
|
877 |
+# make_plot(pos1, smartAlign2(pos1, range = range), range) |
|
878 |
+# |
|
879 |
+# pos1 = rbind(c(1, 8), c(3, 10)) |
|
880 |
+# make_plot(pos1, smartAlign2(pos1, range = range), range) |
|
881 |
+# make_plot(pos1, smartAlign2(pos1, range = range, range_fixed = FALSE), range) |
|
882 |
+# |
|
883 |
+smartAlign2 = function(start, end, range, range_fixed = TRUE) { |
|
884 |
+ |
|
885 |
+ if(missing(end)) { |
|
886 |
+ x1 = start[, 1] |
|
887 |
+ x2 = start[, 2] |
|
888 |
+ } else { |
|
889 |
+ x1 = start |
|
890 |
+ x2 = end |
|
891 |
+ } |
|
892 |
+ |
|
893 |
+ if(missing(range)) { |
|
894 |
+ range = range(c(x1, x2)) |
|
895 |
+ } |
|
896 |
+ |
|
897 |
+ od = order(x1) |
|
898 |
+ rk = rank(x1, ties.method = "random") |
|
899 |
+ x1 = x1[od] |
|
900 |
+ x2 = x2[od] |
|
901 |
+ h = x2 - x1 |
|
902 |
+ |
|
903 |
+ ncluster.before = -1 |
|
904 |
+ ncluster = length(x1) |
|
905 |
+ i_try = 0 |
|
906 |
+ while(ncluster.before != ncluster) { |
|
907 |
+ ncluster.before = ncluster |
|
908 |
+ cluster = rep(0, length(x1)) |
|
909 |
+ i_cluster = 1 |
|
910 |
+ cluster[1] = i_cluster |
|
911 |
+ for(i in seq_along(x1)[-1]) { |
|
912 |
+ # overlap with previous one |
|
913 |
+ if(x1[i] <= x2[i-1]) { # this means x1 should be sorted increasingly |
|
914 |
+ cluster[i] = i_cluster |
|
915 |
+ } else { |
|
916 |
+ i_cluster = i_cluster + 1 |
|
917 |
+ cluster[i] = i_cluster |
|
918 |
+ } |
|
919 |
+ } |
|
920 |
+ ncluster = length(unique(cluster)) |
|
921 |
+ |
|
922 |
+ if(ncluster.before == ncluster && i_try > 0) |