1. ComplexHeatmap
  2. vignettes_old
  3. generate_random_dataset.R
Browse code

remove old vignettes

jokergoo authored on 30/10/2018 12:52:23
Showing 1 changed files
vignettes_old/generate_random_dataset.R
History View file @ b5d9976d
1 1 
deleted file mode 100644
... ... 
@@ -1,117 +0,0 @@
1 
-
2 
-set.seed(888)
3 
-
4 
-type = c(rep("Tumor", 10), rep("Control", 10))
5 
-gender = sample(c("F", "M"), 20, replace = TRUE)
6 
-gender[sample(1:20, 2)] = NA
7 
-age = runif(20, min = 30, max = 80)
8 
-mutation = data.frame(mut1 = sample(c(TRUE, FALSE), 20, p = c(0.2, 0.8), replace = TRUE),
9 
-    mut2 = sample(c(TRUE, FALSE), 20, p = c(0.3, 0.7), replace = TRUE))
10 
-anno = data.frame(type = type, gender = gender, age = age, mutation, stringsAsFactors = FALSE)
11 
-
12 
-######################################
13 
-# generate methylation matrix
14 
-rand_meth = function(k, mean) {
15 
-    (runif(k) - 0.5)*min(c(1-mean), mean) + mean
16 
-}
17 
-
18 
-mean_meth = c(rand_meth(300, 0.3), rand_meth(700, 0.7))
19 
-mat_meth = as.data.frame(lapply(mean_meth, function(m) {
20 
-    if(m < 0.3) {
21 
-        c(rand_meth(10, m), rand_meth(10, m + 0.2))
22 
-    } else if(m > 0.7) {
23 
-        c(rand_meth(10, m), rand_meth(10, m - 0.2))
24 
-    } else {
25 
-        c(rand_meth(10, m), rand_meth(10, m + sample(c(1, -1), 1)*0.2))
26 
-    }
27 
-
28 
-}))
29 
-mat_meth = t(mat_meth)
30 
-rownames(mat_meth) = NULL
31 
-colnames(mat_meth) = paste0("sample", 1:20)
32 
-
33 
-######################################
34 
-# generate directions for methylation
35 
-direction = rowMeans(mat_meth[, 1:10]) - rowMeans(mat_meth[, 11:20])
36 
-direction = ifelse(direction > 0, "hyper", "hypo")
37 
-
38 
-#######################################
39 
-# generate expression matrix
40 
-mat_expr = t(apply(mat_meth, 1, function(x) {
41 
-    x = x + rnorm(length(x), sd = abs(runif(1)-0.5)*0.4 + 0.1)
42 
-    -scale(x)
43 
-}))
44 
-dimnames(mat_expr) = dimnames(mat_meth)
45 
-
46 
-#############################################################
47 
-# matrix for correlation between methylation and expression
48 
-cor_pvalue = -log10(sapply(seq_len(nrow(mat_meth)), function(i) {
49 
-    cor.test(mat_meth[i, ], mat_expr[i, ])$p.value
50 
-}))
51 
-
52 
-#####################################################
53 
-# matrix for types of genes
54 
-gene_type = sample(c("protein_coding", "lincRNA", "microRNA", "psedo-gene", "others"),
55 
-    nrow(mat_meth), replace = TRUE, prob = c(6, 1, 1, 1, 1))
56 
-
57 
-#################################################
58 
-# annotation to genes
59 
-anno_gene = sapply(mean_meth, function(m) {
60 
-    if(m > 0.6) {
61 
-        if(runif(1) < 0.8) return("intragenic")
62 
-    }
63 
-    if(m < 0.4) {
64 
-        if(runif(1) < 0.4) return("TSS")
65 
-    }
66 
-    return("intergenic")
67 
-})
68 
-
69 
-############################################
70 
-# distance to genes
71 
-tss_dist = sapply(mean_meth, function(m) {
72 
-    if(m < 0.3) {
73 
-        if(runif(1) < 0.5) {
74 
-            return(round( (runif(1) - 0.5)*1000 + 500))
75 
-        } else {
76 
-            return(round( (runif(1) - 0.5)*10000 + 500))
77 
-        }
78 
-    } else if(m < 0.6) {
79 
-        if(runif(1) < 0.8) {
80 
-            return(round( (runif(1)-0.5)*100000 + 50000 ))
81 
-        } else {
82 
-            return(round( (runif(1)-0.5)*1000000 + 500000 ))
83 
-        }
84 
-    }
85 
-    return(round( (runif(1) - 0.5)*1000000 + 500000))
86 
-})
87 
-
88 
-
89 
-#######################################
90 
-# annotation to enhancers
91 
-rand_tss = function(m) {
92 
-    if(m < 0.4) {
93 
-        if(runif(1) < 0.25) return(runif(1))
94 
-    } else if (runif(1) < 0.1) {
95 
-        return(runif(1))
96 
-    }
97 
-    return(0)
98 
-}
99 
-rand_enhancer = function(m) {
100 
-    if(m < 0.4) {
101 
-        if(runif(1) < 0.6) return(runif(1))
102 
-    } else if (runif(1) < 0.1) {
103 
-        return(runif(1))
104 
-    }
105 
-    return(0)
106 
-}
107 
-rand_repressive = function(m) {
108 
-    if(m > 0.4) {
109 
-        if(runif(1) < 0.8) return(runif(1))
110 
-    }
111 
-    return(0)
112 
-}
113 
-anno_states = data.frame(
114 
-    tss = sapply(mean_meth, rand_tss),
115 
-    enhancer = sapply(mean_meth, rand_enhancer),
116 
-    rand_repressive = sapply(mean_meth, rand_repressive))
117 
-
Browse code

a backup push

Zuguang Gu authored on 18/09/2018 10:40:29
Showing 1 changed files
vignettes_old/generate_random_dataset.R
History View file @ 30d2c5b2
1 1 
new file mode 100644
... ... 
@@ -0,0 +1,117 @@
1 
+
2 
+set.seed(888)
3 
+
4 
+type = c(rep("Tumor", 10), rep("Control", 10))
5 
+gender = sample(c("F", "M"), 20, replace = TRUE)
6 
+gender[sample(1:20, 2)] = NA
7 
+age = runif(20, min = 30, max = 80)
8 
+mutation = data.frame(mut1 = sample(c(TRUE, FALSE), 20, p = c(0.2, 0.8), replace = TRUE),
9 
+    mut2 = sample(c(TRUE, FALSE), 20, p = c(0.3, 0.7), replace = TRUE))
10 
+anno = data.frame(type = type, gender = gender, age = age, mutation, stringsAsFactors = FALSE)
11 
+
12 
+######################################
13 
+# generate methylation matrix
14 
+rand_meth = function(k, mean) {
15 
+    (runif(k) - 0.5)*min(c(1-mean), mean) + mean
16 
+}
17 
+
18 
+mean_meth = c(rand_meth(300, 0.3), rand_meth(700, 0.7))
19 
+mat_meth = as.data.frame(lapply(mean_meth, function(m) {
20 
+    if(m < 0.3) {
21 
+        c(rand_meth(10, m), rand_meth(10, m + 0.2))
22 
+    } else if(m > 0.7) {
23 
+        c(rand_meth(10, m), rand_meth(10, m - 0.2))
24 
+    } else {
25 
+        c(rand_meth(10, m), rand_meth(10, m + sample(c(1, -1), 1)*0.2))
26 
+    }
27 
+
28 
+}))
29 
+mat_meth = t(mat_meth)
30 
+rownames(mat_meth) = NULL
31 
+colnames(mat_meth) = paste0("sample", 1:20)
32 
+
33 
+######################################
34 
+# generate directions for methylation
35 
+direction = rowMeans(mat_meth[, 1:10]) - rowMeans(mat_meth[, 11:20])
36 
+direction = ifelse(direction > 0, "hyper", "hypo")
37 
+
38 
+#######################################
39 
+# generate expression matrix
40 
+mat_expr = t(apply(mat_meth, 1, function(x) {
41 
+    x = x + rnorm(length(x), sd = abs(runif(1)-0.5)*0.4 + 0.1)
42 
+    -scale(x)
43 
+}))
44 
+dimnames(mat_expr) = dimnames(mat_meth)
45 
+
46 
+#############################################################
47 
+# matrix for correlation between methylation and expression
48 
+cor_pvalue = -log10(sapply(seq_len(nrow(mat_meth)), function(i) {
49 
+    cor.test(mat_meth[i, ], mat_expr[i, ])$p.value
50 
+}))
51 
+
52 
+#####################################################
53 
+# matrix for types of genes
54 
+gene_type = sample(c("protein_coding", "lincRNA", "microRNA", "psedo-gene", "others"),
55 
+    nrow(mat_meth), replace = TRUE, prob = c(6, 1, 1, 1, 1))
56 
+
57 
+#################################################
58 
+# annotation to genes
59 
+anno_gene = sapply(mean_meth, function(m) {
60 
+    if(m > 0.6) {
61 
+        if(runif(1) < 0.8) return("intragenic")
62 
+    }
63 
+    if(m < 0.4) {
64 
+        if(runif(1) < 0.4) return("TSS")
65 
+    }
66 
+    return("intergenic")
67 
+})
68 
+
69 
+############################################
70 
+# distance to genes
71 
+tss_dist = sapply(mean_meth, function(m) {
72 
+    if(m < 0.3) {
73 
+        if(runif(1) < 0.5) {
74 
+            return(round( (runif(1) - 0.5)*1000 + 500))
75 
+        } else {
76 
+            return(round( (runif(1) - 0.5)*10000 + 500))
77 
+        }
78 
+    } else if(m < 0.6) {
79 
+        if(runif(1) < 0.8) {
80 
+            return(round( (runif(1)-0.5)*100000 + 50000 ))
81 
+        } else {
82 
+            return(round( (runif(1)-0.5)*1000000 + 500000 ))
83 
+        }
84 
+    }
85 
+    return(round( (runif(1) - 0.5)*1000000 + 500000))
86 
+})
87 
+
88 
+
89 
+#######################################
90 
+# annotation to enhancers
91 
+rand_tss = function(m) {
92 
+    if(m < 0.4) {
93 
+        if(runif(1) < 0.25) return(runif(1))
94 
+    } else if (runif(1) < 0.1) {
95 
+        return(runif(1))
96 
+    }
97 
+    return(0)
98 
+}
99 
+rand_enhancer = function(m) {
100 
+    if(m < 0.4) {
101 
+        if(runif(1) < 0.6) return(runif(1))
102 
+    } else if (runif(1) < 0.1) {
103 
+        return(runif(1))
104 
+    }
105 
+    return(0)
106 
+}
107 
+rand_repressive = function(m) {
108 
+    if(m > 0.4) {
109 
+        if(runif(1) < 0.8) return(runif(1))
110 
+    }
111 
+    return(0)
112 
+}
113 
+anno_states = data.frame(
114 
+    tss = sapply(mean_meth, rand_tss),
115 
+    enhancer = sapply(mean_meth, rand_enhancer),
116 
+    rand_repressive = sapply(mean_meth, rand_repressive))
117 
+