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
|
#include "SoloReadFeature.h"
#include "serviceFuns.cpp"
#include "SequenceFuns.h"
#include "ReadAnnotations.h"
#include "SoloReadBarcode.h"
class ReadSoloFeatures {
public:
uint32 gene;
vector<uint32> geneMult;
vector<array<uint64,2>> sj;
bool sjAnnot;
uint32 indAnnotTr; //index of the annotated transcript
Transcript **alignOut;
};
uint32 outputReadCB(fstream *streamOut, const uint64 iRead, const int32 featureType, SoloReadBarcode &soloBar,
const ReadSoloFeatures &reFe, const ReadAnnotations &readAnnot, const SoloReadFlagClass &readFlag);
void SoloReadFeature::record(SoloReadBarcode &soloBar, uint nTr, Transcript **alignOut, uint64 iRead, ReadAnnotations &readAnnot)
{
if (pSolo.type==0)
return;
if (pSolo.readStatsYes[featureType]) {//readFlag
if (nTr==1) {
readFlag.setBit(readFlag.genomeU);
} else if (nTr>1) {
readFlag.setBit(readFlag.genomeM);
};
for (uint64 itr=0; itr<nTr; itr++) {
if (P.pGe.chrSet.mito.count(alignOut[itr]->Chr) == 1) {
readFlag.setBit(readFlag.mito);
};
};
switch (readAnnot.annotFeatures[featureType].ovType) {
case ReadAnnotFeature::overlapTypes::exonic :
case ReadAnnotFeature::overlapTypes::exonic50p :
readFlag.setBit(readFlag.exonic);
break;
case ReadAnnotFeature::overlapTypes::intronic :
readFlag.setBit(readFlag.intronic);
break;
case ReadAnnotFeature::overlapTypes::exonicAS :
case ReadAnnotFeature::overlapTypes::exonic50pAS :
readFlag.setBit(readFlag.exonicAS);
break;
case ReadAnnotFeature::overlapTypes::intronicAS :
readFlag.setBit(readFlag.intronicAS);
};
if (soloBar.cbMatch<0 && pSolo.cbWLyes) {//no CB match in the WL
if (readAnnot.annotFeatures[featureType].fSet.size()==1) {
readFlag.setBit(readFlag.featureU);
} else if (readAnnot.annotFeatures[featureType].fSet.size()>1){
readFlag.setBit(readFlag.featureM);
};
readFlag.setBit(readFlag.cbMatch);//this will counts reads with no CB match
readFlag.countsAddNoCB();
};
};
if (soloBar.cbMatch<0)
return;
ReadSoloFeatures reFe;
reFe.alignOut=alignOut;
reFe.indAnnotTr = 0;
uint32 nFeat=0; //number of features in this read (could be >1 for SJs)
if (nTr==0) {//unmapped
stats.V[stats.noUnmapped]++;
} else {
switch (featureType) {
case SoloFeatureTypes::Gene :
case SoloFeatureTypes::GeneFull :
case SoloFeatureTypes::GeneFull_Ex50pAS :
case SoloFeatureTypes::GeneFull_ExonOverIntron :
{
auto *readGe = &readAnnot.annotFeatures[featureType].fSet;
if (soloBar.pSolo.type==soloBar.pSolo.SoloTypes::SmartSeq) {
for (int32 itr=nTr-1; itr>=0; itr--) {
if (readAnnot.annotFeatures[featureType].fAlign[itr].size() > 0) {
reFe.indAnnotTr = itr;
break;//indAnnotTr is the first genic align, and is used for Smart-seq deduplication, below in outputReadCB()
//TODO this is ambiguous for reads that map to the same gene multiple times. *Last* align is chosen to match the previous version.
};
};
};
if (readGe->size()==0) {//check genes
stats.V[stats.noNoFeature]++;//no gene
} else if (readGe->size()>1) {
stats.V[stats.MultiFeature]++;//multi-gene reads
readFlag.setBit(readFlag.featureM);
if (nTr>1)
stats.V[stats.subMultiFeatureMultiGenomic]++;//multigene caused by multimapper
if (pSolo.multiMap.yes.multi) {//output multimappers
reFe.geneMult.resize(readGe->size());
uint32 ii=0;
for (auto &g: *readGe) {//set high bit to mark multimappers
reFe.geneMult[ii] = g | geneMultMark;
++ii;
};
nFeat = outputReadCB(streamReads, iRead, featureType, soloBar, reFe, readAnnot, readFlag);
};
} else {//unique-gene reads
reFe.gene = *readGe->begin();
readFlag.setBit(readFlag.featureU);
nFeat = outputReadCB(streamReads, (readIndexYes ? iRead : (uint64)-1), featureType, soloBar, reFe, readAnnot, readFlag);
};
//debug
//{
// if (SoloFeatureTypes::Gene==featureType && ((readFlag.checkBit(readFlag.featureM)==1|readFlag.checkBit(readFlag.featureU)==1) != readFlag.checkBit(readFlag.exonic) ))
// cout << iRead<<" "<<readFlag.checkBit(readFlag.featureM)<<" "<<readFlag.checkBit(readFlag.featureU)<<" "<<readFlag.checkBit(readFlag.exonic)<<endl;
//};
};
break;
case SoloFeatureTypes::SJ :
if (nTr>1) {//reject all multimapping reads
stats.V[stats.subMultiFeatureMultiGenomic]++;
stats.V[stats.MultiFeature]++;
//} else if (readAnnot.geneConcordant.size()>1){//for SJs, still check genes, no feature if multi-gene
// stats.V[stats.MultiFeature]++;
} else {//one gene or no gene
alignOut[0]->extractSpliceJunctions(reFe.sj, reFe.sjAnnot);
//if ( reFe.sj.empty() || (reFe.sjAnnot && readAnnot.geneConcordant.size()==0) ) {//no junctions, or annotated junction but no gene (i.e. read does not fully match transcript)
if ( reFe.sj.empty() ) {
stats.V[stats.noNoFeature]++;
} else {//good junction
readFlag.setBit(readFlag.featureU);
nFeat = outputReadCB(streamReads, (readIndexYes ? iRead : (uint64)-1), featureType, soloBar, reFe, readAnnot, readFlag);
};
};
break;
case SoloFeatureTypes::Transcript3p :
if (readAnnot.transcriptConcordant.size()==0 || soloBar.cbMatch>1) {//do not record ambiguous CB
stats.V[stats.noNoFeature]++;
} else {
nFeat = outputReadCB(streamReads, iRead, featureType, soloBar, reFe, readAnnot, readFlag);
};
if (readAnnot.transcriptConcordant.size()==1 && readAnnot.transcriptConcordant[0][1] < transcriptDistCount.size()) {
//read maps to one transcript - use for distTTS distribution function
transcriptDistCount[readAnnot.transcriptConcordant[0][1]]++;
};
break;
case SoloFeatureTypes::Velocyto :
//different record: iRead, nTr, tr1, type1, tr2, type2 ...
if (readAnnot.trVelocytoType.size()>0) {//otherwise, no gene
sort(readAnnot.trVelocytoType.begin(), readAnnot.trVelocytoType.end(),
[](const trTypeStruct &t1, const trTypeStruct &t2) {return t1.tr < t2.tr;});
*streamReads << iRead <<' '<< readAnnot.trVelocytoType.size();
for (auto &tt: readAnnot.trVelocytoType)
*streamReads <<' '<< tt.tr <<' '<< (uint32) tt.type;
*streamReads <<'\n';
nFeat=1;
} else {
stats.V[stats.noNoFeature]++;
};
break; //no need to go with downstream processing
};//switch (featureType)
};//if (nTr==0)
if ( nFeat==0 && (readInfoYes | pSolo.readStatsYes[featureType]) ) {//no feature, but readInfo requested
outputReadCB(streamReads, iRead, (uint32)-1, soloBar, reFe, readAnnot, readFlag);
};
if (nFeat==0)
return; //no need to record the number of reads per CB
if (pSolo.cbWLyes) {//WL
for (auto &cbi : soloBar.cbMatchInd)
cbReadCount[cbi] += nFeat;
} else {//no WL
cbReadCountMap[soloBar.cbMatchInd[0]] += nFeat;
};
/*//if we wanted to record all barcodes, even those without features
if (!pSolo.cbWLyes) {//no WL
cbReadCountMap[soloBar.cbMatchInd[0]] += nFeat;
} else if (nFeat>0){//WL && nFeat>0
for (auto &cbi : soloBar.cbMatchInd)
cbReadCount[cbi] += nFeat;
};
*/
return;
};
uint32 outputReadCB(fstream *streamOut, const uint64 iRead, const int32 featureType, SoloReadBarcode &soloBar,
const ReadSoloFeatures &reFe, const ReadAnnotations &readAnnot, const SoloReadFlagClass &readFlag)
{
/*format of the temp output file
* UMI [iRead] type feature* cbMatchString
* 0=exact match, 1=one non-exact match, 2=multipe non-exact matches
* gene or sj[0] sj[1]
* CB or nCB {CB Qual, ...}
*/
if (soloBar.pSolo.type==soloBar.pSolo.SoloTypes::SmartSeq && featureType!=-1) {//need to calculate "UMI" from align start/end
soloBar.umiB=reFe.alignOut[reFe.indAnnotTr]->chrStartLengthExtended();
};
uint64 nout=1;
switch (featureType) {
case -1 :
//no feature, output for readInfo
*streamOut << soloBar.umiB <<' '<< iRead <<' '<< readFlag.flag <<' '<< -1 <<' '<< soloBar.cbMatch <<' '<< soloBar.cbMatchString <<'\n';
break;
case SoloFeatureTypes::Gene :
case SoloFeatureTypes::GeneFull :
case SoloFeatureTypes::GeneFull_Ex50pAS :
case SoloFeatureTypes::GeneFull_ExonOverIntron :
if (reFe.geneMult.size()==0) {
//just gene id
*streamOut << soloBar.umiB <<' ';//UMI
if ( iRead != (uint64)-1 )
*streamOut << iRead <<' '<< readFlag.flag <<' ';//iRead
*streamOut << reFe.gene <<' '<< soloBar.cbMatch <<' '<< soloBar.cbMatchString <<'\n';;
} else {
for (auto &g : reFe.geneMult) {
*streamOut << soloBar.umiB <<' ';//UMI
*streamOut << iRead <<' '<< readFlag.flag <<' ';//iRead is always output for multiGene
*streamOut << g <<' '<< soloBar.cbMatch <<' '<< soloBar.cbMatchString <<'\n';;
};
nout = reFe.geneMult.size();
};
break;
case SoloFeatureTypes::SJ :
//sj - two numbers, multiple sjs per read
for (auto &sj : reFe.sj) {
*streamOut << soloBar.umiB <<' ';//UMI
if ( iRead != (uint64)-1 )
*streamOut << iRead <<' '<< readFlag.flag <<' ';//iRead
*streamOut << sj[0] <<' '<< sj[1] <<' '<< soloBar.cbMatch <<' '<< soloBar.cbMatchString <<'\n' << flush;
};
nout=reFe.sj.size();
break;
case SoloFeatureTypes::Transcript3p :
//transcript,distToTTS structure
*streamOut << soloBar.cbMatchString <<' ';
*streamOut << soloBar.umiB <<' ';
*streamOut << readAnnot.transcriptConcordant.size();
for (auto &tt: readAnnot.transcriptConcordant) {
*streamOut <<' '<< tt[0] <<' '<< tt[1];
};
if ( iRead != (uint64)-1 )
*streamOut <<' '<< iRead;//iRead
*streamOut <<'\n';
nout=1;
break;
}; //switch (featureType)
return nout;
};
|
