forked from weitinglin/Bioconductor-Coursera
-
Notifications
You must be signed in to change notification settings - Fork 0
/
week 2
324 lines (264 loc) · 10.9 KB
/
week 2
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
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
##Bioconductor for genomic data
#quiz 2
#20151019 weitinglin
#Introduction
#examine Epigenomics Roadmap data on the human H1 stem cell line;the Roadmap id number is "E003". We will use data on build"hg19",for peak will use narrowPeak.focus on chromosome 22
source("http://www.bioconductor.org/biocLite.R")
biocLite("BSgenome.Hsapiens.UCSC.hg19")
#Q1
library(BSgenome)
library(AnnotationHub)
library(BSgenome.Hsapiens.UCSC.hg19)
available.genomes()
seqlengths(BSgenome.Hsapiens.UCSC.hg19)
seqnames(BSgenome.Hsapiens.UCSC.hg19)
BSgenome.Hsapiens.UCSC.hg19$chr22
A<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"A",as.prob=TRUE)
C<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"C",as.prob=TRUE)
T<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"T",as.prob=TRUE)
G<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"G",as.prob=TRUE)
N<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"N",as.prob=TRUE)
nA<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"A",as.prob=FALSE)
nC<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"C",as.prob=FALSE)
nT<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"T",as.prob=FALSE)
nG<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"G",as.prob=FALSE)
nN<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"N",as.prob=FALSE)
fA<-nA/(nA+nC+nT+nG)
fC<-nC/(nA+nC+nT+nG)
fT<-nT/(nA+nC+nT+nG)
fG<-nG/(nA+nC+nT+nG)
nC*nG/(nA+nC+nT+nG)
A_frequency<-(A)
A+C+T+G-N
(C+G)/(A+C+T+G)
#0.47988
#Q2
library(AnnotationHub)
ah=AnnotationHub()
ah_1=query(ah,c("H3K27me3","narrowPeak"))
ah_1=query(ah_1,c("E003"))
ah_1=ah_1[["AH29892"]]
seqlevels(ah_1)
ah_chr22=keepSeqlevels(ah_1,c("chr22"))
ah_chr22
trim(ah_chr22)
v1<-Views(Hsapiens, ah_chr22)
letterFrequency(v1,"GC",as.prob=TRUE)
mean(letterFrequency(v1,"GC",as.prob=TRUE))
head(v1)
#0.004467924
#Q3 the correlation between GC content and SignalValue
matrix<-data.frame(letterFrequency(v1,"GC",as.prob=TRUE),ah_chr22$signalValue)
cor(matrix)
cor(letterFrequency(v1,"GC",as.prob=TRUE),ah_chr22$signalValue)
sv=mcols(v1)
#Q4 to cope with bigwig data
library(rtracklayer)
ah_chip<-query(ah,c("fc.signal"))
ah_chip<-query(ah_chip,c("H3K27me3"))
ah_chip<-query(ah_chip,c("E003"))
ah_chip<-ah_chip[["AH32033"]]
class(ah_chip)
ranges(ah_chr22)
ah_chip.bw<-import(ah_chip,which=ah_chr22)
chr22<-GRanges(seqnames="chr22",ranges=IRanges(start=1,end=51304566))
ah_chip.bw<-import(ah_chip,which=chr22,as="Rle")
table(table(ah_chip.bw$chr22))
class(ah_chip.bw$chr22)
ah_chip.bw.view<-Views(ah_chip.bw$chr22,start=start(ah_chr22),end=end(ah_chr22)) #what I had got from here?
mean(ah_chip.bw.view)
cor(mean(ah_chip.bw.view),ah_chr22$signalValue)
head(runLength(ah_chip.bw$chr22))
ah_chip.bw$chr22
#Q5 how many bases on chr22, which fc.signal >=1
try<-ah_chip.bw.view[1]
width(ah_chip.bw.view[[2]])
runLength(ah_chip.bw.view[[2]])
class(runValue(ah_chip.bw.view[[2]]))
ah_chip.bw.view
objects(ah_chip.bw.view)
class(ah_chip.bw.view)
sort(runValue(try))
head(runValue(try))
ah_chip.bw.view>=1
sum(ah_chip.bw$chr22>=1) #這寫法也太神來之筆了吧!
sum(sum(ah_chip.bw.view>=1))
class(ah_chip.bw$chr22)
class(ah_chip.bw.view)
?Rle
sum(runValue(ah_chip.bw$chr22)>=1) #這樣寫也行
ah_chip.bw.view
#Q6 examine the differential expression on E003(<0.5) and E055(>2)
#acquire the annotation data
ah_E055=query(ah,c("H3K27me3","narrowPeak"))
ah_E055=query(ah_E055,c("E055"))
ah_E055<-ah_E055[["AH30313"]]
#acquire the ChIP data bigwig
ah_E055_fc=query(ah,c("E055","fc.signal"))
ah_E055_fc=query(ah_E055_fc,c("H3K27me3"))
ah_E055_fc=ah_E055_fc[["AH32470"]]
ah_E055_fcimport
chr22<-GRanges(seqnames="chr22",ranges=IRanges(start=1,end=51304566))
ah_E055_fcimport<-import(ah_E055_fc,which=chr22,as="Rle")
ah_E055_fcimport$chr22 #from E055
ah_chip.bw$chr22 #from E003
ah_E055_fcimport$chr22
ah_E055_fcimport[chr22]
#slice
E055_ir<-slice(ah_E055_fcimport$chr22,lower=2)
E003_ir<-slice(ah_chip.bw$chr22,upper=0.5)
#turn to Ranges
E055_ir_ranges=as(E055_ir,"IRanges")
E003_ir_ranges=as(E003_ir,"IRanges")
E055_ir_ranges
#intersect
comb_ir=intersect(E055_ir_ranges,E003_ir_ranges)
sum(width(comb_ir))
#1869937
#Q7 What is the average observed-to-expected ratio of CpG dinucleotide for CpG islands on chromosome 22.
CG<-letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"CG",as.prob=TRUE)
letterFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,"GC",as.prob=TRUE)
CG/(fC*fG) #可能GC的值跟我們想要的“GC”連在一起的值不同
#alphabetFrequency,letterFrequency : compute the frequency of ALL Characters
#dinucleotideFrequency,trinucleotideFrequency,oligonucleotideFrequency
#letterFrequencyInSlidingView: letter frequencies, but in sliding views along the string
#consensusMatrix:consensus matrix; all a position weight matrix
?dinucleotideFrequency
?letterFrequencyInSlidingView
#使用dinucleotideFrequency
dinucleotideFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22)
class(dinucleotideFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22))
dinucleotideFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22)["AA"]
AA<-dinucleotideFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22)["AA"]
A_A<-dinucleotideFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22)
sum(A_A[2])
CG_f<-dinucleotideFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22)["CG"]
#使用oligonucleotideFrequency
oligonucleotideFrequency(BSgenome.Hsapiens.UCSC.hg19$chr22,width=3)
#使用letterFrequencyInSlidingView
## letterFrequencyInSlidingView()
## ---------------------------------------------------------------------
data(yeastSEQCHR1)
x <- DNAString(yeastSEQCHR1)
view.width <- 48
letters <- c("A", "CG")
two_columns <- letterFrequencyInSlidingView(x, view.width, letters)
head(two_columns)
tail(two_columns)
three_columns <- letterFrequencyInSlidingView(x, view.width, letters, OR=0)
head(three_columns)
tail(three_columns)
stopifnot(identical(two_columns[ , "C|G"],
three_columns[ , "C"] + three_columns[ , "G"]))
## Note that, alternatively, 'three_columns' can also be obtained by
## creating the views on 'x' (as a Views object) and by calling
## alphabetFrequency() on it. But, of course, that is be *much* less
## efficient (both, in terms of memory and speed) than using
## letterFrequencyInSlidingView():
v <- Views(x, start=seq_len(length(x) - view.width + 1), width=view.width)
v
three_columns2 <- alphabetFrequency(v, baseOnly=TRUE)[ , c("A", "C", "G")]
stopifnot(identical(three_columns2, three_columns))
## Set the width of the view to length(x) to get the global frequencies:
letterFrequencyInSlidingView(x, letters="ACGTN", view.width=length(x), OR=0)
sum(letterFrequencyInSlidingView(BSgenome.Hsapiens.UCSC.hg19$chr22,letters="CG", view.width=2))
slide_CG<-sum((letterFrequencyInSlidingView(BSgenome.Hsapiens.UCSC.hg19$chr22,letters="CG", view.width=2))!=0)
CG_f
nC+nG
expected_CG
CG_f/expected_CG # 用as.numberic避免溢位
expected_CG<-(as.numeric(nC)*as.numeric(nG))/as.numeric(nA+nC+nT+nG+nN)
expected_CG<-(as.numeric(nC)*as.numeric(nG))/as.numeric(nA+nC+nT+nG) #比較有無加上nN的差別,似乎增加不多。
##此法不可行,另闢其他作法,找有CpG的database
##仔細思考題意,應該是要專注在CpG islands
CpG_islands<-query(ah,c("hg19","CpG"))
CpG_islands<-subset(CpG_islands,species=="Homo sapiens")
CpG_islands<-CpG_islands[["AH5086"]]
chromosome<-c(paste("chr",1:22))
CpG_islands_s<-split(CpG_islands,seqnames(CpG_islands))
CpG_islands_22<-CpG_islands_s$chr22
CpG_islands_22
head((start(CpG_islands_22)))
CpG_islands_vi<-Views(BSgenome.Hsapiens.UCSC.hg19$chr22,start=start(CpG_islands_22),end=end(CpG_islands_22))
CpG_G<-letterFrequency(CpG_islands_vi,"G")
CpG_C<-letterFrequency(CpG_islands_vi,"C")
CG<-(dinucleotideFrequency(CpG_islands_vi)) # 變成matrix
CG[,7]
CpG_G*CpG_C
CpG_G
CpG_C
201*206
CpG_islands_vi
width(CpG_islands_22) #不能用sum喔 觀念不同,待會再平均
mean((CG[,7]*width(CpG_islands_22))/(CpG_G*CpG_C))
#0.8340929
BSgenome.Hsapiens.UCSC.hg19$chr22
chr22_hsapiens<-subset(BSgenome.Hsapiens.UCSC.hg19,se)
#Q8 How many TATA BOX on chromosome 22
library(Biostrings)
TATA<-DNAString("TATAAA")
class(BSgenome.Hsapiens.UCSC.hg19$chr22)
rematch_TATA<-matchPattern(reverseComplement(TATA),BSgenome.Hsapiens.UCSC.hg19$chr22)
match_TATA<-matchPattern(TATA,BSgenome.Hsapiens.UCSC.hg19$chr22)
length(rematch_TATA)+length(match_TATA)
#27263
#收DNAstring,但這做法只有search其中一個strand
vmatch_TATA<-vmatchPattern(TATA,BSgenome.Hsapiens.UCSC.hg19)
vmatch_TATA<-vmatchPattern(TATA,BSgenome.Hsapiens.UCSC.hg19$chr22)
vmatch_TATAs<-split(vmatch_TATA,seqnames(vmatch_TATA))
vmatch_TATAs$chr22
#27263
#Q9:How manay promoters of transcripts on chr22 containg a coding sequence, contain a TATA box on the same strand as the transcripts? use TxDb.Hsapiens.UCSC.hg19.knownGene
#Q10
library(GenomicFeatures)
library(TxDb.Hsapiens.UCSC.hg19.knownGene)
txdb<-TxDb.Hsapiens.UCSC.hg19.knownGene
head(genes(txdb))
class(txdb)
seqlevels(txdb,force=TRE)
library(GenomicFeatures)
library(TxDb.Hsapiens.UCSC.hg19.knownGene)
txdb <- TxDb.Hsapiens.UCSC.hg19.knownGene
seqlevels(txdb, force=TRUE) <- c("chr22")
gr <- GRanges(seqnames = "chr22", ranges = IRanges(start = 1, end = 52330658))
gr.trans.chr22 <- subsetByOverlaps(transcripts(txdb), gr, ignore.strand = TRUE)
gr.trans.chr22
length(gr.trans.chr22)
gr.prom <- promoters(gr.trans.chr22, upstream = 900, downstream = 100)
class(gr.prom )
class(gr.trans.chr22)
gr.trans.chr22
gr.prom
gr.cds<-cds(txdb)
length(gr.cds)
?cds
#acquire the sequency
final_gr_prom_sequence<-Views(BSgenome.Hsapiens.UCSC.hg19$chr22,start=start(promindex),end=end(promindex))
gr_cds_sequence<-Views(BSgenome.Hsapiens.UCSC.hg19$chr22,start=start(cdsindex),end=end(cdsindex))
trans_promoter_chr22_gr.cds
promindex<-gr.prom[unique(queryHits(findOverlaps(gr.prom,gr.cds)))]
cdsindex<-gr.trans.chr22[unique(subjectHits(findOverlaps(gr.prom,gr.trans.chr22)))] #wrong
ffinal_gr_prom_sequence<-gr.prom[unique(queryHits(findOverlaps(prom_cds_trans,gr.prom)))]
head(final_gr_prom_sequence)
reverseComplement(TATA)
length(matchPattern(TATA,final_gr_prom_sequence))+length(matchPattern(reverseComplement(TATA),final_gr_prom_sequence))
length(matchPattern(TATA,gr_prom_sequence))+length(matchPattern(reverseComplement(TATA),gr_prom_sequence))
# use bsapply
class(trans_promoter_chr22_gr.cds)
rans_promoter_chr22_sequence_rTATA<-matchPattern(TATA,trans_promoter_chr22_gr.cds)
rans_promoter_chr22_sequence_rTATA
trans_promoter_chr22_gr.cds
class(trans_promoter_chr22_sequence)
trans_promoter_chr22_sequence
countPattern(TATA,trans_promoter_chr22_gr.cds)
rans_promoter_chr22_sequence_TATA<-vmatchPattern(TATA,trans_promoter_chr22_gr.cds)
rans_promoter_chr22_sequence_TATA
rans_promoter_chr22_sequence_rTATA<-matchPattern(TATA,trans_promoter_chr22_gr.cds)
length(rans_promoter_chr22_sequence_TATA)+length(rans_promoter_chr22_sequence_rTATA)
length(rans_promoter_chr22_sequence_TATA)
#178
tl.chr22 <- transcriptLengths(txdb, with.cds_len = TRUE) #rtn df
tl.chr22 <- tl.chr22[tl.chr22$cds_len > 0,]
trans.eval <- gr.prom[mcols(gr.prom)$tx_id %in% tl.chr22$tx_id]
sum(coverage(trans.eval) > 1)
tl.chr22