Data-Analysis-of-Variant-Functional-Effects/fg_manuscript_postprocessing.R at main · KChen-lab/Data-Analysis-of-Variant-Functional-Effects · GitHub

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
# R version 4.2.2 (2022-10-31) -- "Innocent and Trusting"
# R Core Team (2022). R: A language and environment for statistical computing. R Foundation for Statistical
# Computing, Vienna, Austria. URL https://www.R-project.org/.
# Copyright (C) 2022 The R Foundation for Statistical Computing
# Platform: x86_64-apple-darwin17.0 (64-bit)

# Transvar: HG19, UCSC
# Input genome	hg19
# open-cravat	2.3.0
# Gene mapper	UCSC hg38 Gene Mapper (1.10.4)

library(dplyr)
library(ggplot2)
library(ggpubr)

#### results from cravat ####
cravat_output <-
  read.table(
    'cravat/cravat_output_hg19_ucsc.tsv',
    header = T,
    skip = 6,
    sep = '\t'
  )
cravat.headers = colnames(cravat_output)
# expand chasmplus and vest4
cravat.headers.select = cravat.headers[c(1:12, 15, 20, 21,
                                         26, 27, 45:49, 54:64, 66, 67)]

new.headers = c(
  cravat.headers[1:12],
  'CGC_class',
  'CHASMplus_pval',
  'CHASMplus_score',
  'CScape_score',
  'DANN_score',
  'MA_rank_score',
  'MA_functional_impact',
  'MT_rank_score',
  'MT_prediction',
  'MP_rank_score',
  'PhD_SNPg_prediction',
  'PhD_SNPg_score',
  'PP2_hdiv_prediction',
  'PP2_hvar_prediction',
  'PP2_hdiv_rank_score',
  'PP2_hvar_rank_score',
  'PROVEAN_rank_score',
  'PROVEAN_prediction',
  'SIFT_prediction',
  'SIFT_rank_score',
  'Siphy_rank_score',
  'VEST4_score',
  'VEST4_pval'
)
cravat.output.select <- cravat_output[, cravat.headers.select]
colnames(cravat.output.select) <- new.headers
CGI.aa <-
  full_join(CGI, transvar_aa_map, by = c('Alteration' = 'input'))
CGI.cravat = full_join(CGI.aa, cravat.output.select, by = c("id" = "Samples"))

# driver/passenger prediction
# CHASMplus_score: probability of being driver for missense mutations
# VEST4_pval: p-value of being pathogenic for all nonsilent consequence types
CGI.cravat.chasmvest = CGI.cravat %>% # filter(Oncogenic.Summary=="") %>%
  select(Oncogenic.Summary,
         CHASMplus_score,
         CScape_score,
         DANN_score,
         VEST4_pval)

p1 = ggplot(CGI.cravat.chasmvest,
            aes(x = Oncogenic.Summary, y = CHASMplus_score)) + geom_boxplot()
p2 = ggplot(CGI.cravat.chasmvest,
            aes(x = Oncogenic.Summary, y = CScape_score)) + geom_boxplot()
p3 = ggplot(CGI.cravat.chasmvest, aes(x = Oncogenic.Summary, y = DANN_score)) + geom_boxplot()
p4 = ggplot(CGI.cravat.chasmvest, aes(x = Oncogenic.Summary, y = VEST4_pval)) + geom_boxplot()

ggarrange(p1, p2, p3, p4, nrow = 2, ncol = 2)
# chasm, cscape, and vest 4 are better than dann; vest4 is the best

# prediction
CGI.cravat$VEST4_pred <-
  ifelse(CGI.cravat$VEST4_pval < 0.05, 'driver (predicted)', 'passenger')
CGI.cravat$CHASMplus_pred <-
  ifelse(CGI.cravat$CHASMplus_pval < 0.05,
         'driver (predicted)',
         'passenger')
CGI.cravat$CScape_pred <-
  ifelse(CGI.cravat$CScape_score > 0.5,
         'driver (predicted)',
         'passenger')

CGI.cravat.result = CGI.cravat %>% select(
  id,
  Alteration,
  p.aa1infer,
  p.check,
  p.aa3infer,
  Oncogenic.Summary,
  Chrom,
  Position,
  Ref_Base,
  Alt_Base,
  Gene,
  Transcript,
  Sequence_Ontology,
  c.infer,
  cDNA_change,
  Protein_Change,
  VEST4_pred,
  CHASMplus_pred,
  CScape_pred
)

# QC on protein change position
ppos1 <- strsplit(CGI.cravat.result$Alteration, ":")
ppos1 <-
  sapply(strsplit(CGI.cravat.result$Alteration, ":"), tail, 1)
ppos1 <- as.numeric(gsub(".*?([0-9]+).*", "\\1", ppos1))

ppos2 <-
  as.numeric(gsub(".*?([0-9]+).*", "\\1", CGI.cravat.result$Protein_Change))
CGI.cravat.result$ppos.check = ppos1 == ppos2

# remove unmatched multiple mapping with at least one matched
map.count <-
  CGI.cravat.result %>% group_by(id) %>% count() %>% filter(n > 1)
map.true.count <-
  CGI.cravat.result %>% filter(ppos.check) %>% group_by(id) %>% count()
map.count.tofilter <- intersect(map.count$id, map.true.count$id)

CGI.cravat.result$ppos.filter <-
  CGI.cravat$id %in% map.count.tofilter
CGI.cravat.result$ppos.filter <-
  CGI.cravat.result$ppos.filter & CGI.cravat.result$ppos.check == F
CGI.cravat.result2 <- CGI.cravat.result %>% filter(ppos.filter == F)

CGI.cravat.result2$pcheck_transvar_cravat <-
  CGI.cravat.result2$p.aa3infer == CGI.cravat.result2$Protein_Change
CGI.cravat.result2 %>% group_by(pcheck_transvar_cravat) %>% count()
CGI.cravat.result2$ccheck_transvar_cravat <-
  CGI.cravat.result2$c.infer == CGI.cravat.result2$cDNA_change
CGI.cravat.result2 %>% group_by(ccheck_transvar_cravat) %>% count()

CGI.cravat.result2 %>% group_by(ccheck_transvar_cravat, pcheck_transvar_cravat) %>% count()

write.csv(CGI.cravat.result2,
          'cravat/cravat_hg19_ucsc_analysis_result.csv',
          row.names = F)

CGI.cravat.result2 %>% filter(!ccheck_transvar_cravat |
                                !pcheck_transvar_cravat) %>% write.csv('cravat/ucsc_hg19_match_check_descrepancies.csv')