-
Notifications
You must be signed in to change notification settings - Fork 0
/
LegacyCode.R
190 lines (185 loc) · 6.04 KB
/
LegacyCode.R
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
### This function identifies the sequence to which each paired read should map
mapPairedReads = function(F1, F2, uniqueKmers, tree = NULL) {
L = nrow(F1)
stopifnot(L == nrow(F2))
Maps = vector("list", L)
M = length(uniqueKmers)
N = nchar(uniqueKmers[[1]][1])
Fracs = rep(0, M)
allNames = sapply(names(uniqueKmers), function(x) {unlist(strsplit(x, SEP))})
redNames = unlist(allNames[sapply(allNames, length) == 1])
names(Fracs) = redNames
for (ind in 1:L) {
cur1 = getKmers(F1[ind,], N, FALSE)
cur2 = getKmers(F2[ind,], N, FALSE)
if (!is.null(tree)) {
node = rootNode(tree)
kids = children(tree, node)
cur = c()
while (length(kids) > 0) {
stopifnot(length(kids) == 2)
leftLabels = labels(tree)[descendants(tree, kids[1], "tips")]
leftLabel = paste(leftLabels, collapse = SEP)
rightLabels = labels(tree)[descendants(tree, kids[2], "tips")]
rightLabel = paste(rightLabels, collapse = SEP)
leftBranch1 = any(uniqueKmers[[leftLabel]] %in% cur1)
rightBranch1 = any(uniqueKmers[[rightLabel]] %in% cur1)
leftBranch2 = any(uniqueKmers[[leftLabel]] %in% cur2)
rightBranch2 = any(uniqueKmers[[rightLabel]] %in% cur2)
anyLeft = leftBranch1 || leftBranch2
anyRight = rightBranch1 || rightBranch2
stopifnot(!(anyLeft && anyRight))
if (anyLeft) {
node = kids[1]
}
else if (anyRight) {
node = kids[2]
}
else {
break
}
kids = children(tree, node)
}
if (length(kids) == 0) {
cur = labels(tree)[node]
}
Maps[[ind]] = cur
Fracs[cur] = Fracs[cur] + 1
}
else {
set1 = which(sapply(1:M, function(x) {any(uniqueKmers[[x]] %in% cur1)}))
set2 = which(sapply(1:M, function(x) {any(uniqueKmers[[x]] %in% cur2)}))
if (length(setdiff(set1, set2)) == 0) {
cur = set2
}
else if (length(setdiff(set2, set1)) == 0) {
cur = set1
}
else {
cur = c()
}
stopifnot(length(cur) <= 1)
Maps[[ind]] = cur
Fracs[cur] = Fracs[cur] + 1
}
}
print(paste(round(sum(Fracs) / L * 100, 2), "% of the reads got assigned"))
Fracs = Fracs[Fracs > 0]
Fracs = Fracs/sum(Fracs)
output = list(Maps, Fracs)
output
}
### This function is a wrapper function for a case of only known strains present, error-free reads.
wrapperKnown = function(baseFilename, redAlign, N, byRead, tree) {
correctFractions = parseFilenameFracs(baseFilename)
File1 = paste0(baseFilename, "R1.fq")
if (!File1 %in% list.files()) {
File1 = paste0(File1, ".txt")
}
File2 = paste0(baseFilename, "R2.fq")
if (!File2 %in% list.files()) {
File2 = paste0(File2, ".txt")
}
F1 = parseF(File1, keepNames = TRUE)
F2 = parseF(File2, keepNames = TRUE)
print("Constructing the de Bruijn graph")
G0 = constructDBGraph(rbind(F1, F2), N)
Res = identifyPaths(redAlign, G0[[1]], N, G0[[3]], remove = FALSE, maxMissingFrac = MMF, count = TRUE)
goodStrains = names(Res)[!sapply(Res, is.null)]
redTree = NULL
contrastList = NULL
if (!is.na(tree) && length(goodStrains) > 2) {
redTree = subset(phylo4(tree), tips.include = goodStrains)
contrastList = computeContrasts(redTree)
}
Qred = getUniqueKmers(redAlign[goodStrains, , drop = FALSE], N, contrastList = contrastList, intersect = (!is.na(tree) && !byRead))
if (byRead) {
C = mapPairedReads(F1, F2, Qred[[2]], tree = redTree)[[2]]
}
else {
C = getMeanCoverage(Res, Qred, goodStrains, save = SAVE, baseFilename = baseFilename, tree = redTree)
}
output = list(correctFractions, C)
output
}
### Unfinished function!
findMaxCompatible = function(allMaps, index, K) {
curPos = lapply(allMaps, function(x) {x[2, x[1,] == index]})
goodPos = sapply(curPos, function(x) {length(x) > 0})
curPos[!goodPos] = NA
if (length(goodPos) > 1) {
curPos = unlist(curPos)
curBest = length(goodPos) ### THIS IS WRONG, FIX LATER!
### FIND LONGEST PROGRESSION WITH DIFFERENCE K!
}
else {
curBest = 1
}
curBest
}
### This function tries to map reads based on a table of unique K-mers
mapReads = function(readTable, uniqueKmers, allKmers, maxError = 1) {
K = nchar(uniqueKmers[[1]][1])
N = length(uniqueKmers)
R = nrow(readTable)
L = ncol(readTable)
numKmers = L - K + 1
map = rep(0, R)
names(map) = rownames(readTable)
for (ind in 1:R) {
curRead = readTable[ind,]
curKmers = getKmers(curRead, K)
if (is.null(curKmers)) {
next
}
for (index in 1:N) {
mappedKmers = which(curKmers %in% uniqueKmers[[index]])
if (length(mappedKmers) > 0) {
firstIndex = mappedKmers[1]
firstMapped = curKmers[firstIndex]
curSequence = allKmers[[index]]
mapPos = which(curSequence == firstMapped)
if (mapPos - firstIndex + numKmers <= length(curSequence)) {
curMapped = (curSequence[mapPos - firstIndex + (1:numKmers)] == curKmers)
if (sum(!curMapped) <= maxError * K) {
map[ind] = index
}
}
break
}
}
}
map
}
### This function takes in a reference sequence and a read and returns the best place to align a given read to it and a score
alignRead = function(refSequence, read, method = "window") {
if (method == "window") {
L0 = length(refSequence)
L1 = length(read)
deltaL = L0 - L1 + 1
scores = rep(NA, deltaL)
freqs = table(c(refSequence, read))
freqs = freqs[ALPHABET]
for (pos in 1:deltaL) {
Table = table(refSequence[pos + (0:(L1 - 1))], read)
Table = Table[ALPHABET, ALPHABET]
scores[pos] = distanceFormula(Table, freqs)
}
bestScore = max(scores)
bestPos = which.max(scores)
}
result = list(position = bestPos, score = bestScore)
}
### This function takes in a reference database and a read and returns the vector of supports for each sequence in the database
alignReadToDB = function(DB, read, threshold, ranking = "best", tol = 0.01) {
M = nrow(DB)
supports = rep(0, M)
scores = rep(0, M)
for (ind in 1:M) {
curAlign = alignRead(DB[ind,], read)
scores[ind] = curAlign$score
}
bestRefs = which(scores >= max(score) - tol)
supports[bestRefs] = 1
supports
}