|
a |
|
b/Figures/clonevol/clonevol_P33_141118.R |
|
|
1 |
library(dplyr) |
|
|
2 |
library(plyr) |
|
|
3 |
library(clonevol) |
|
|
4 |
library(fishplot) |
|
|
5 |
library(reshape) |
|
|
6 |
|
|
|
7 |
####coverage: from reseq BAM files |
|
|
8 |
####step1: Using mpileup to extract coverage for each site. (Prepared 27/10/18), default parameter |
|
|
9 |
|
|
|
10 |
####step2: Prepare input for pyclone, using segments from ASCAT |
|
|
11 |
|
|
|
12 |
####step3: Run Pyclone twice. The first run is to identify founding cluster, |
|
|
13 |
#### the second run adding --tumour_content based on the cellular prevalence for each biopsy (29/10/18) |
|
|
14 |
|
|
|
15 |
###step4: using clonevol to build evolution tree and model (in this script) (30/10/18) |
|
|
16 |
|
|
|
17 |
### using Fishplot to visulaize |
|
|
18 |
|
|
|
19 |
### using Fishplot to visulaize |
|
|
20 |
|
|
|
21 |
###P33, 2 biopsies, hist_transform |
|
|
22 |
#T1:LN_neck |
|
|
23 |
#T2:LN_supraclav_left |
|
|
24 |
|
|
|
25 |
|
|
|
26 |
|
|
|
27 |
setwd("G:/FL_resequncing/FL_exome_final/fl_latest/11_pyclone/BED_bam/counts_new/pyclone_output_301018") |
|
|
28 |
|
|
|
29 |
pyclone_out<- read.table(file="ouput_pyclone/output_200X_i2/P33_200X_i2/tables/loci.tsv",sep="\t",header=T) |
|
|
30 |
|
|
|
31 |
mutations<-cast(pyclone_out[,1:4], mutation_id~sample_id, value="cellular_prevalence") |
|
|
32 |
id<-unique(pyclone_out[, c(1,3)] ) |
|
|
33 |
|
|
|
34 |
P_case<- merge(id, mutations,by="mutation_id") |
|
|
35 |
|
|
|
36 |
vafs = data.frame(cluster=P_case$cluster_id, |
|
|
37 |
T1_vaf=(P_case$`GCF0150-0033-T01`/2)*100, |
|
|
38 |
T2_vaf=(P_case$`GCF0150-0033-T02`/2)*100, |
|
|
39 |
stringsAsFactors=F) |
|
|
40 |
|
|
|
41 |
vafs$cluster<- vafs$cluster+1 |
|
|
42 |
|
|
|
43 |
###needs manully check density plot to identify which cluster is founding cluster |
|
|
44 |
##clonevol requires founding cluster=1 |
|
|
45 |
|
|
|
46 |
max_id<- max(vafs$cluster)+1 |
|
|
47 |
|
|
|
48 |
vafs$cluster[vafs$cluster==1]<-max_id |
|
|
49 |
vafs$cluster[vafs$cluster==5]<-1 |
|
|
50 |
vafs$cluster[vafs$cluster==max_id]<-5 |
|
|
51 |
|
|
|
52 |
|
|
|
53 |
samples<-c("P33_1","P33_2") |
|
|
54 |
samples1<-c("P33_1\nPrimary","P33_2\nTransformed") |
|
|
55 |
samples2<-c("P33_1\nPrimary\nLN_neck","P33_2\nTransformed\nLN_supraclav") |
|
|
56 |
|
|
|
57 |
|
|
|
58 |
|
|
|
59 |
|
|
|
60 |
|
|
|
61 |
names(vafs)[2:3] = samples |
|
|
62 |
##step 2: run infer.clonal.models, run twice: 1. include all cluster. |
|
|
63 |
###########2. manual review density plot and exlcude cluster with small number of mutations |
|
|
64 |
##step 2: run infer.clonal.models, run twice: 1. include all cluster. |
|
|
65 |
###########2. manual review density plot and exlcude cluster with small number of mutations |
|
|
66 |
|
|
|
67 |
|
|
|
68 |
dir.create("./clonevol/P33") |
|
|
69 |
## |
|
|
70 |
|
|
|
71 |
|
|
|
72 |
##first: use all clusters, no consensus models |
|
|
73 |
|
|
|
74 |
#vafs$P31_3[vafs$cluster=="1"]<-vafs$P31_3[vafs$cluster=="1"]+ 0.5 |
|
|
75 |
#vafs$P27_2[vafs$cluster=="1"]<-vafs$P27_2[vafs$cluster=="1"]+ 0.5 |
|
|
76 |
#vafs$P27_3[vafs$cluster=="3"]<-vafs$P27_3[vafs$cluster=="3"]+1 |
|
|
77 |
|
|
|
78 |
res = infer.clonal.models(variants=vafs, cluster.col.name="cluster", vaf.col.names=samples, |
|
|
79 |
|
|
|
80 |
subclonal.test="bootstrap", subclonal.test.model="non-parametric", |
|
|
81 |
founding.cluster=1,ignore.clusters = c(3,6,9:13), |
|
|
82 |
cluster.center="mean", num.boots=1000, |
|
|
83 |
min.cluster.vaf=0.01, sum.p=0.01, alpha=0.01) |
|
|
84 |
|
|
|
85 |
|
|
|
86 |
|
|
|
87 |
vafs_used<- subset(vafs, !cluster %in% c(3,6,9:13)) |
|
|
88 |
|
|
|
89 |
vafs_used$cluster[vafs_used$cluster==8]<-3 |
|
|
90 |
vafs_used$cluster[vafs_used$cluster==7]<-6 |
|
|
91 |
|
|
|
92 |
res = infer.clonal.models(variants=vafs_used, cluster.col.name="cluster", vaf.col.names=samples, |
|
|
93 |
|
|
|
94 |
subclonal.test="bootstrap", subclonal.test.model="non-parametric", |
|
|
95 |
founding.cluster=1, |
|
|
96 |
cluster.center="mean", num.boots=1000, |
|
|
97 |
min.cluster.vaf=0.01, sum.p=0.01, alpha=0.01) |
|
|
98 |
|
|
|
99 |
|
|
|
100 |
|
|
|
101 |
|
|
|
102 |
res<-convert.consensus.tree.clone.to.branch(res, branch.scale = 'sqrt') |
|
|
103 |
|
|
|
104 |
pdf("./clonevol/P33/P33_trees.pdf", useDingbats = FALSE) |
|
|
105 |
plot.all.trees.clone.as.branch(res, branch.width = 0.5, node.size = 1, node.label.size = 0.5) |
|
|
106 |
dev.off() |
|
|
107 |
|
|
|
108 |
|
|
|
109 |
plot.clonal.models(res, |
|
|
110 |
# box plot parameters |
|
|
111 |
box.plot = TRUE, |
|
|
112 |
fancy.boxplot = TRUE, |
|
|
113 |
fancy.variant.boxplot.highlight = 'is.driver', |
|
|
114 |
fancy.variant.boxplot.highlight.shape = 21, |
|
|
115 |
fancy.variant.boxplot.highlight.fill.color = 'red', |
|
|
116 |
fancy.variant.boxplot.highlight.color = 'black', |
|
|
117 |
fancy.variant.boxplot.highlight.note.col.name = 'gene', |
|
|
118 |
fancy.variant.boxplot.highlight.note.color = 'blue', |
|
|
119 |
fancy.variant.boxplot.highlight.note.size = 2, |
|
|
120 |
fancy.variant.boxplot.jitter.alpha = 1, |
|
|
121 |
fancy.variant.boxplot.jitter.center.color = 'grey50', |
|
|
122 |
fancy.variant.boxplot.base_size = 12, |
|
|
123 |
fancy.variant.boxplot.plot.margin = 1, |
|
|
124 |
fancy.variant.boxplot.vaf.suffix = '.VAF', |
|
|
125 |
# bell plot parameters |
|
|
126 |
clone.shape = 'bell', |
|
|
127 |
bell.event = TRUE, |
|
|
128 |
bell.event.label.color = 'blue', |
|
|
129 |
bell.event.label.angle = 60, |
|
|
130 |
clone.time.step.scale = 1, |
|
|
131 |
bell.curve.step = 2, |
|
|
132 |
# node-based consensus tree parameters |
|
|
133 |
merged.tree.plot = TRUE, |
|
|
134 |
tree.node.label.split.character = NULL, |
|
|
135 |
tree.node.shape = 'circle', |
|
|
136 |
tree.node.size = 30, |
|
|
137 |
tree.node.text.size = 0.5, |
|
|
138 |
merged.tree.node.size.scale = 1.25, |
|
|
139 |
merged.tree.node.text.size.scale = 2, |
|
|
140 |
merged.tree.cell.frac.ci = FALSE, |
|
|
141 |
# branch-based consensus tree parameters |
|
|
142 |
merged.tree.clone.as.branch = TRUE, |
|
|
143 |
mtcab.event.sep.char = ',', |
|
|
144 |
mtcab.branch.text.size = 1, |
|
|
145 |
mtcab.branch.width = 0.75, |
|
|
146 |
mtcab.node.size = 3, |
|
|
147 |
mtcab.node.label.size = 1, |
|
|
148 |
mtcab.node.text.size = 1.5, |
|
|
149 |
# cellular population parameters |
|
|
150 |
cell.plot = TRUE, |
|
|
151 |
num.cells = 100, |
|
|
152 |
cell.border.size = 0.25, |
|
|
153 |
cell.border.color = 'black', |
|
|
154 |
clone.grouping = 'horizontal', |
|
|
155 |
#meta-parameters |
|
|
156 |
scale.monoclonal.cell.frac = TRUE, |
|
|
157 |
show.score = FALSE, |
|
|
158 |
cell.frac.ci = TRUE, |
|
|
159 |
disable.cell.frac = FALSE, |
|
|
160 |
# output figure parameters |
|
|
161 |
out.dir = './clonevol/P33/', |
|
|
162 |
out.format = 'pdf', |
|
|
163 |
overwrite.output = TRUE, |
|
|
164 |
width = 10, |
|
|
165 |
height = 4, |
|
|
166 |
# vector of width scales for each panel from left to right |
|
|
167 |
panel.widths = c(1.5,2.5,1.5,2.5,2)) |
|
|
168 |
|
|
|
169 |
###removing cell.frac annotation |
|
|
170 |
|
|
|
171 |
plot.clonal.models(res, |
|
|
172 |
# box plot parameters |
|
|
173 |
box.plot = TRUE, |
|
|
174 |
fancy.boxplot = TRUE, |
|
|
175 |
fancy.variant.boxplot.highlight = 'is.driver', |
|
|
176 |
fancy.variant.boxplot.highlight.shape = 21, |
|
|
177 |
fancy.variant.boxplot.highlight.fill.color = 'red', |
|
|
178 |
fancy.variant.boxplot.highlight.color = 'black', |
|
|
179 |
fancy.variant.boxplot.highlight.note.col.name = 'gene', |
|
|
180 |
fancy.variant.boxplot.highlight.note.color = 'blue', |
|
|
181 |
fancy.variant.boxplot.highlight.note.size = 2, |
|
|
182 |
fancy.variant.boxplot.jitter.alpha = 1, |
|
|
183 |
fancy.variant.boxplot.jitter.center.color = 'grey50', |
|
|
184 |
fancy.variant.boxplot.base_size = 12, |
|
|
185 |
fancy.variant.boxplot.plot.margin = 1, |
|
|
186 |
fancy.variant.boxplot.vaf.suffix = '.VAF', |
|
|
187 |
# bell plot parameters |
|
|
188 |
clone.shape = 'bell', |
|
|
189 |
bell.event = TRUE, |
|
|
190 |
bell.event.label.color = 'blue', |
|
|
191 |
bell.event.label.angle = 60, |
|
|
192 |
clone.time.step.scale = 1, |
|
|
193 |
bell.curve.step = 2, |
|
|
194 |
# node-based consensus tree parameters |
|
|
195 |
merged.tree.plot = TRUE, |
|
|
196 |
tree.node.label.split.character = NULL, |
|
|
197 |
tree.node.shape = 'circle', |
|
|
198 |
tree.node.size = 30, |
|
|
199 |
tree.node.text.size = 0.5, |
|
|
200 |
merged.tree.node.size.scale = 1.25, |
|
|
201 |
merged.tree.node.text.size.scale = 2, |
|
|
202 |
merged.tree.cell.frac.ci = FALSE, |
|
|
203 |
# branch-based consensus tree parameters |
|
|
204 |
merged.tree.clone.as.branch = TRUE, |
|
|
205 |
mtcab.event.sep.char = ',', |
|
|
206 |
mtcab.branch.text.size = 1, |
|
|
207 |
mtcab.branch.width = 0.75, |
|
|
208 |
mtcab.node.size = 3, |
|
|
209 |
mtcab.node.label.size = 1, |
|
|
210 |
mtcab.node.text.size = 1.5, |
|
|
211 |
# cellular population parameters |
|
|
212 |
cell.plot = TRUE, |
|
|
213 |
num.cells = 100, |
|
|
214 |
cell.border.size = 0.25, |
|
|
215 |
cell.border.color = 'black', |
|
|
216 |
clone.grouping = 'horizontal', |
|
|
217 |
#meta-parameters |
|
|
218 |
scale.monoclonal.cell.frac = TRUE, |
|
|
219 |
show.score = FALSE, |
|
|
220 |
cell.frac.ci = TRUE, |
|
|
221 |
disable.cell.frac = TRUE, |
|
|
222 |
# output figure parameters |
|
|
223 |
out.dir = './clonevol/P33/', |
|
|
224 |
out.format = 'pdf', |
|
|
225 |
overwrite.output = TRUE, |
|
|
226 |
width = 10, |
|
|
227 |
height = 4, |
|
|
228 |
# vector of width scales for each panel from left to right |
|
|
229 |
panel.widths = c(1.5,2.5,1.5,2.5,2)) |
|
|
230 |
|
|
|
231 |
|
|
|
232 |
##generating fish plot |
|
|
233 |
f<- generateFishplotInputs(results = res) |
|
|
234 |
fishes=createFishPlotObjects(f) |
|
|
235 |
|
|
|
236 |
|
|
|
237 |
|
|
|
238 |
pdf('./clonevol/P33/P33_fish_200x_pyclone_anno_loc.pdf', width=14, height=7) |
|
|
239 |
for (i in 1:length(fishes)){ |
|
|
240 |
|
|
|
241 |
fish = layoutClones(fishes[[i]]) |
|
|
242 |
fish = setCol(fish,f$clonevol.clone.colors) |
|
|
243 |
fishPlot(fish,shape="spline", title.btm="P1", cex.title=0.7,cex.vlab = 1.4, |
|
|
244 |
vlines=seq(1, length(samples2)), vlab=samples2, pad.left=0.5) |
|
|
245 |
} |
|
|
246 |
dev.off() |
|
|
247 |
|
|
|
248 |
|
|
|
249 |
|
|
|
250 |
pdf("./clonevol/P33/P33_box.pdf", width=3, height=3,useDingbats = FALSE, title='') |
|
|
251 |
pp<-plot.variant.clusters(vafs_used, |
|
|
252 |
cluster.col.name = 'cluster', |
|
|
253 |
show.cluster.size = FALSE, |
|
|
254 |
cluster.size.text.color = 'blue', |
|
|
255 |
vaf.col.names = samples, |
|
|
256 |
vaf.limits = 70, |
|
|
257 |
sample.title.size = 20, |
|
|
258 |
violin = FALSE, |
|
|
259 |
box = FALSE, |
|
|
260 |
jitter = TRUE, |
|
|
261 |
jitter.shape = 1, |
|
|
262 |
|
|
|
263 |
jitter.size = 3, |
|
|
264 |
jitter.alpha = 1, |
|
|
265 |
jitter.center.method = 'median', |
|
|
266 |
jitter.center.size = 1, |
|
|
267 |
jitter.center.color = 'darkgray', |
|
|
268 |
jitter.center.display.value = 'none', |
|
|
269 |
highlight = 'is.driver', |
|
|
270 |
highlight.shape = 21, |
|
|
271 |
highlight.color = 'blue', |
|
|
272 |
highlight.fill.color = 'green', |
|
|
273 |
highlight.note.col.name = 'gene', |
|
|
274 |
highlight.note.size = 2, |
|
|
275 |
order.by.total.vaf = FALSE) |
|
|
276 |
|
|
|
277 |
dev.off() |
|
|
278 |
|
|
|
279 |
|
|
|
280 |
plot.pairwise(vafs_used, col.names = samples, |
|
|
281 |
out.prefix = './clonevol/P33/P33_variants.pairwise.plot') |
|
|
282 |
|
|
|
283 |
|
|
|
284 |
pdf('./clonevol/P33/P33_flow.pdf') |
|
|
285 |
plot.cluster.flow(vafs, vaf.col.names = samples, |
|
|
286 |
sample.names = c('Primary', 'Transformed')) |
|
|
287 |
dev.off() |
|
|
288 |
|
|
|
289 |
|
|
|
290 |
####checking coverage |
|
|
291 |
Pcase<-do.call("rbind", lapply( list.files("input_pyclone_271018_newpara/200X/GCF0150-0033-N01_200X/",full=TRUE), |
|
|
292 |
read.table, header=TRUE, sep="\t")) |
|
|
293 |
|
|
|
294 |
|
|
|
295 |
|
|
|
296 |
|
|
|
297 |
######## |
|
|
298 |
#min (dp) =min(c1inP4_1$var_counts+c1inP4_1$ref_counts)=273 |
|
|
299 |
|
|
|
300 |
#max (dp) =max(c1inP4_1$var_counts+c1inP4_1$ref_counts)=648 |
|
|
301 |
#median (dp) =median(c1inP4_1$var_counts+c1inP4_1$ref_counts)=380 |
|
|
302 |
#mean (dp) =mean(c1inP4_1$var_counts+c1inP4_1$ref_counts)=417 |
|
|
303 |
|
|
|
304 |
library(ggplot2) |
|
|
305 |
|
|
|
306 |
|
|
|
307 |
|
|
|
308 |
pdf("clonevol/P33/coverage.pdf") |
|
|
309 |
ggplot(Pcase, aes(x=(var_counts+ref_counts)))+ |
|
|
310 |
geom_histogram(position="dodge")+ |
|
|
311 |
facet_grid(~sample) |
|
|
312 |
|
|
|
313 |
dev.off() |
|
|
314 |
|
|
|
315 |
save.image("G:/FL_resequncing/FL_exome_final/fl_latest/11_pyclone/BED_bam/counts_new/pyclone_output_301018/P33.RData") |
|
|
316 |
|
|
|
317 |
|