-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathreport.rmd
255 lines (216 loc) · 9.93 KB
/
report.rmd
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
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
---
title: "NeoFlow report"
date: "`r Sys.Date()`"
output:
BiocStyle::html_document:
toc_float: true
params:
input_dir: ""
prefix: ""
out_dir: ""
min_n_sample_name: 50
mhc_filter: 150
vignette: >
%\VignetteIndexEntry{NeoFlow report}
%\VignetteEngine{knitr::rmarkdown}
%\VignetteKeywords{Mass Spectrometry, Proteomics, Genomics, neoantigen }
%\VignetteEncoding{UTF-8}
---
```{r setup, include=FALSE}
library(knitr)
library(png)
library(kableExtra)
library(tidyr)
library(dplyr)
library(stringr)
library(formattable)
#library(DT)
knitr::opts_chunk$set(echo = FALSE)
```
```{r Import data, echo=FALSE}
input_dir <- params$input_dir
use_log <- c(1,1)
prefix <- params$prefix
out_dir <- params$out_dir
min_n_sample_name <- params$min_n_sample_name
mhc_filter <- params$mhc_filter
```
```{r echo=FALSE}
knitr::opts_chunk$set(cache = TRUE, warning = FALSE,
message = FALSE, cache.lazy = FALSE)
```
```{r echo=FALSE}
################################################################################
############################# functions ########################################
# Generate a summary file which contain everything we need for downstream analysis
# This file contains the following information:
# mutation information at DNA level
# mutation information at protein level
# HLA type
# epitope
# binding affinity
# variant peptide information
load_data=function(neoflow_dir="./",out_dir="./"){
dir.create(out_dir,showWarnings = FALSE,recursive = TRUE)
if(str_detect(neoflow_dir,pattern = "^s3:")){
neoflow_dir_local <- paste0(out_dir,"/neoflow_dir")
dir.create(neoflow_dir_local,showWarnings = FALSE,recursive = TRUE)
s3_cmd <- paste0("aws s3 sync ",neoflow_dir,"neoantigen_prediction/ ",neoflow_dir_local)
system(s3_cmd)
neoflow_dir <- neoflow_dir_local
}
fs <- list.files(path = neoflow_dir,pattern = ".tsv",include.dirs = TRUE,recursive = TRUE,full.names = TRUE)
dd <- lapply(fs, function(k){
a <- data.table::fread(k,colClasses=list(character=c("Chr","Start","End","Ref","Alt"))) %>%
select(Chr,Start,End,Ref,Alt,Variant_Type,Gene,mRNA,Variant_Start,Variant_End,AA_before,AA_after,Neoepitope,HLA_type,netMHCpan_binding_affinity_nM,protein_var_evidence_pep) %>%
distinct()
nm <- dirname(k) %>% basename() %>% str_split(pattern = "_") %>% unlist()
a$sample <- nm[length(nm)]
return(a)
}) %>% bind_rows()
dd$pep_evidence <- ifelse(dd$protein_var_evidence_pep=="-","No","Yes")
return(dd)
}
## generate data for overview barplot
generate_summary_file=function(x,mhc_filter=150,prefix="test"){
################################################################################
## output summary information
res <- x
res_filtered <- res %>% filter(netMHCpan_binding_affinity_nM <= mhc_filter)
#cat("Summary information:\n")
output_stat <- list()
output_stat$total_somatic_mutations <- res %>%
select(Chr,Start,End,Ref,Alt,sample) %>%
distinct() %>% group_by(sample) %>%
summarise(total_somatic_mutations=n())
output_stat$neoantigen_somatic_mutations <- res_filtered %>%
select(Chr,Start,End,Ref,Alt,sample) %>%
distinct() %>% group_by(sample) %>%
summarise(neoantigen_somatic_mutations=n())
output_stat$neoantigen_somatic_mutations_with_pro_var <- res_filtered %>%
filter(protein_var_evidence_pep!="-") %>%
select(Chr,Start,End,Ref,Alt,sample) %>%
distinct() %>% group_by(sample) %>%
summarise(neoantigen_somatic_mutations_with_pro_var=n())
output_stat$neoantigen <- res_filtered %>%
select(Neoepitope,sample) %>%
distinct() %>% group_by(sample) %>%
summarise(neoantigen=n())
output_stat$neoantigen_with_pro_var <- res_filtered %>%
filter(protein_var_evidence_pep!="-") %>%
select(Neoepitope,sample) %>%
distinct() %>% group_by(sample) %>%
summarise(neoantigen_with_pro_var=n())
res_stat <- merge(output_stat$total_somatic_mutations,output_stat$neoantigen_somatic_mutations,
by="sample",all.x=TRUE) %>%
merge(output_stat$neoantigen_somatic_mutations_with_pro_var,by="sample",all.x=TRUE) %>%
merge(output_stat$neoantigen,by="sample",all.x=TRUE) %>%
merge(output_stat$neoantigen_with_pro_var,by="sample",all.x=TRUE)
for(i in 2:ncol(res_stat)){
y <- res_stat[,i]
y[is.na(y)] <- 0
res_stat[,i] <- y
}
return(res_stat)
}
## overview barplot
plot_hist=function(x,use_log,fig,type="somatic",min_n_sample_name=50,mhc_filter=150){
plot_data_path <- paste(dirname(fig),"/hist_",type,"_plot_data.rda",sep="")
save(x,use_log,fig,type,min_n_sample_name,file=plot_data_path)
a <- x %>% arrange(desc(total_somatic_mutations))
n_samples <- nrow(a)
png(fig,width = 1300,height = 500,res = 150)
library(RColorBrewer)
par(mfrow=c(4,1))
par(mar=c(0,4,1,1),las=2,cex.axis=1.2)
barplot(a$total_somatic_mutations,names.arg = "",col="#8FBC94",log=ifelse(use_log[1]==1,"y",""),ylim = c(min(a$total_somatic_mutations),1.3*max(a$total_somatic_mutations)));graphics::box()
text(x=par("usr")[2]*0.98,y=max(a$total_somatic_mutations)**0.95,
labels=ifelse(type=="somatic","# nonsynonymous somatic mutations","# fusion genes"),
cex=1.5,pos=2)
par(mar=c(0,4,0,1),las=2,cex.axis=1.2)
barplot(a$neoantigen_somatic_mutations,col="gray",names.arg = "",log=ifelse(use_log[2]==1,"y",""),ylim = c(min(a$neoantigen_somatic_mutations),1.3*max(a$neoantigen_somatic_mutations)));graphics::box()
text(x=par("usr")[2]*0.98,y=max(a$neoantigen_somatic_mutations)**0.95,
labels=ifelse(type=="somatic",
paste("# nonsynonymous somatic mutations with neoantigen candidates (binding affinity < ",mhc_filter,"nM)",sep=""),
paste("# fusion genes with neoantigen candidates (binding affinity < ",mhc_filter,"nM)",sep="")),
cex=1.5,pos=2)
par(mar=c(0,4,0,1),cex.axis=1.2,las=2)
barplot(a$neoantigen_somatic_mutations_with_pro_var,names.arg = "",col="lightblue",ylim = c(min(a$neoantigen_somatic_mutations_with_pro_var),1.3*max(a$neoantigen_somatic_mutations_with_pro_var)));graphics::box()
text(x=par("usr")[2]*0.98,y=max(a$neoantigen_somatic_mutations_with_pro_var)**0.95,
labels=ifelse(type=="somatic",
paste("# nonsynonymous somatic mutations with neoantigen candidates (binding affinity < ",mhc_filter,"nM) \n+ protein evidence",sep=""),
paste("# fusion genes with neoantigen candidates (binding affinity < ",mhc_filter,"nM) + protein evidence",sep="")),
cex=1.5,pos=2)
par(mar=c(3,4,0,1),cex.axis=1,mgp=c(2,1,0))
if(nrow(a)<=min_n_sample_name){
barplot(a$neoantigen_with_pro_var,
names.arg = a$sample,
cex.names=0.8,
las=2,
col="red",
ylim = c(0,1.3*max(a$neoantigen_with_pro_var)),xlab="",cex.lab=1.2)
}else{
barplot(a$neoantigen_with_pro_var,
names.arg = "",
las=2,
col="red",
ylim = c(0,1.3*max(a$neoantigen_with_pro_var)),xlab="Sample",cex.lab=1.2)
}
graphics::box()
n_samples_with_protein <- sum(a$neoantigen_with_pro_var>=1)
text(x=par("usr")[2]*0.98,y=max(a$neoantigen_with_pro_var)*0.95,
labels=paste("neoepitope ",n_samples_with_protein,"/",n_samples,sep = ""),
cex=1.5,pos=2)
dev.off()
}
## variant peptide identification summary
```
# Introduction
```{r load_data, echo=FALSE,warning=FALSE,message=FALSE,cache=TRUE}
raw_data <- load_data(input_dir,out_dir = out_dir)
```
# Neoantigen discovery result overview
```{r overview fig, echo=FALSE, fig.align='center', results='asis'}
sample_total <- NA
if(nrow(raw_data %>% filter(Variant_Type!="fusion")) >= 1){
res1 <- raw_data %>% filter(Variant_Type!="fusion") %>% generate_summary_file(mhc_filter=mhc_filter)
fig <- paste(out_dir,"/",prefix,"_somatic_mutation.png",sep = "")
res1 %>% plot_hist(use_log = use_log,fig=fig,min_n_sample_name=min_n_sample_name,mhc_filter=mhc_filter)
sample_total <- res1$sample %>% unique()
knitr::include_graphics(fig %>% normalizePath)
cat("Somatic mutation derived neoantigen")
cat("![](",fig,")")
}
if(nrow(raw_data %>% filter(Variant_Type=="fusion")) >= 1){
res2 <- raw_data %>% filter(Variant_Type=="fusion") %>% generate_summary_file(mhc_filter=mhc_filter)
fig <- paste(out_dir,"/",prefix,"_fusion.png",sep = "")
res2 %>% plot_hist(use_log = c(0,0),fig=paste(prefix,"_fusion.png",sep = ""),type = "fusion",min_n_sample_name=min_n_sample_name,mhc_filter=mhc_filter)
sample_total <- c(sample_total,res2$sample) %>% unique()
knitr::include_graphics(fig %>% normalizePath)
cat("Gene fusion derived neoantigen")
cat("![](",fig,")")
}
```
# Neoantigen discovery result detail
```{r o_table, echo=FALSE}
sample_somatic_mutation <- res1 %>% filter(neoantigen_somatic_mutations_with_pro_var>=1)
if(nrow(raw_data %>% filter(Variant_Type=="fusion")) >= 1){
sample_fusion <- res2 %>% filter(neoantigen_somatic_mutations_with_pro_var>=1)
odt <- data.frame(Item=c("Total samples",
"Total samples with somatic protein evidence",
"Total samples with fusion protein evidence",
"Total samples after combine two evidences"),
Value=c(sample_total %>% length,
nrow(sample_somatic_mutation),
nrow(sample_fusion),
c(sample_somatic_mutation$sample,sample_fusion$sample) %>% unique() %>% length))
}else{
odt <- data.frame(Item=c("Total samples",
"Total samples with somatic protein evidence"),
Value=c(sample_total %>% length,
nrow(sample_somatic_mutation)))
}
kable(odt,"html",escape = FALSE) %>%
kable_styling(bootstrap_options = "striped", full_width = F) %>%
scroll_box(height = "400px",width = "100%")
```