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
325
326
327
328
329
330
331
332
333
334
335
|
#include "SoloFeature.h"
#include "streamFuns.h"
#include "TimeFunctions.h"
#include "serviceFuns.cpp"
#include <unordered_map>
#include "SoloCommon.h"
#include <bitset>
#define def_MarkNoColor (uint32) -1
inline int funCompareSolo1 (const void *a, const void *b) {
uint32 *va= (uint32*) a;
uint32 *vb= (uint32*) b;
if (va[1]>vb[1]) {
return 1;
} else if (va[1]<vb[1]) {
return -1;
} else if (va[0]>vb[0]){
return 1;
} else if (va[0]<vb[0]){
return -1;
} else {
return 0;
};
};
void SoloFeature::collapseUMIall(uint32 iCB, uint32 *umiArray)
{
uint32 *rGU=rCBp[iCB];
uint32 rN=nReadPerCB[iCB];
qsort(rGU,rN,rguStride*sizeof(uint32),funCompareNumbers<uint32>); //sort by gene index
//compact reads per gene
uint32 gid1=-1;//current gID
uint32 nGenes=0; //number of genes
uint32 *gID = new uint32[min(featuresNumber,rN)+1]; //gene IDs
uint32 *gReadS = new uint32[min(featuresNumber,rN)+1]; //start of gene reads TODO: allocate this array in the 2nd half of rGU
for (uint32 iR=0; iR<rN*rguStride; iR+=rguStride) {
if (rGU[iR+rguG]!=gid1) {//record gene boundary
gReadS[nGenes]=iR;
gid1=rGU[iR+rguG];
gID[nGenes]=gid1;
++nGenes;
};
};
gReadS[nGenes]=rguStride*rN;//so that gReadS[nGenes]-gReadS[nGenes-1] is the number of reads for nGenes, see below in qsort
//unordered_map<uint32, uint32> umiMaxGeneCount;//for each umi, max counts of reads per gene
unordered_map <uintUMI, unordered_map<uint32,uint32>> umiGeneHash, umiGeneHash0;
//UMI //Gene //Count
if (pSolo.umiFiltering.MultiGeneUMI) {
for (uint32 iR=0; iR<rN*rguStride; iR+=rguStride) {
umiGeneHash[rGU[iR+1]][rGU[iR]]++;
};
for (auto &iu : umiGeneHash) {//loop over all UMIs
if (iu.second.size()==1)
continue;
uint32 maxu=0;
for (auto &ig : iu.second) {//loop over genes for a given UMI
if (maxu<ig.second)
maxu=ig.second; //find gene with maximum count
};
if (maxu==1)
maxu=2;//to kill UMIs with 1 read to one gene, 1 read to another gene
for (auto &ig : iu.second) {
if (maxu>ig.second)
ig.second=0; //kills Gene with read count *strictly* < maximum count
};
};
};
vector<unordered_map <uintUMI,uintUMI>> umiCorrected(nGenes);
if (countCellGeneUMI.size() < countCellGeneUMIindex[iCB] + nGenes*countMatStride)
countCellGeneUMI.resize((countCellGeneUMI.size() + nGenes*countMatStride )*2);//allocated vector too small
nGenePerCB[iCB]=0;
nUMIperCB[iCB]=0;
countCellGeneUMIindex[iCB+1]=countCellGeneUMIindex[iCB];
/////////////////////////////////////////////
/////////// main cycle over genes
for (uint32 iG=0; iG<nGenes; iG++) {//collapse UMIs for each gene
uint32 *rGU1=rGU+gReadS[iG];
uint32 nR0 = (gReadS[iG+1]-gReadS[iG])/rguStride; //total number of reads
if (nR0==0)
continue; //no reads - this should not happen?
qsort(rGU1, nR0, rguStride*sizeof(uint32), funCompareTypeShift<uint32,rguU>);
//exact collapse
uint32 iR1=-umiArrayStride; //number of distinct UMIs for this gene
uint32 u1=-1;
for (uint32 iR=rguU; iR<gReadS[iG+1]-gReadS[iG]; iR+=rguStride) {//count and collapse identical UMIs
if (pSolo.umiFiltering.MultiGeneUMI && umiGeneHash[rGU1[iR]][gID[iG]]==0) {//multigene UMI is not recorded
if ( pSolo.umiDedup.typeMain != UMIdedup::typeI::NoDedup ) //for NoDedup, the UMI filtering is not done
rGU1[iR] = (uintUMI) -1; //mark multigene UMI, so that UB tag will be set to -
continue;
};
if (rGU1[iR]!=u1) {
iR1 += umiArrayStride;
u1=rGU1[iR];
umiArray[iR1]=u1;
umiArray[iR1+1]=0;
};
umiArray[iR1+1]++;
//if ( umiArray[iR1+1]>nRumiMax) nRumiMax=umiArray[iR1+1];
};
uint32 nU0 = (iR1+umiArrayStride)/umiArrayStride;//number of UMIs after simple exact collapse
if (pSolo.umiFiltering.MultiGeneUMI_CR) {
if (nU0==0)
continue; //nothing to count
for (uint64 iu=0; iu<nU0*umiArrayStride; iu+=umiArrayStride) {
umiGeneHash0[umiArray[iu+0]][iG]+=umiArray[iu+1];//this sums read counts over UMIs that were collapsed
};
umiArrayCorrect_CR(nU0, umiArray, readInfo.size()>0, false, umiCorrected[iG]);
for (uint64 iu=0; iu<nU0*umiArrayStride; iu+=umiArrayStride) {//just fill the umiGeneHash - will calculate UMI counts later
umiGeneHash[umiArray[iu+2]][iG]+=umiArray[iu+1];//this sums read counts over UMIs that were collapsed
};
continue; //done with MultiGeneUMI_CR, readInfo will be filled later
};
if (pSolo.umiDedup.yes.NoDedup)
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + pSolo.umiDedup.countInd.NoDedup] = nR0;
if (nU0>0) {//otherwise no need to count
if (pSolo.umiDedup.yes.Exact)
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + pSolo.umiDedup.countInd.Exact] = nU0;
if (pSolo.umiDedup.yes.CR)
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + pSolo.umiDedup.countInd.CR] =
umiArrayCorrect_CR(nU0, umiArray, readInfo.size()>0 && pSolo.umiDedup.typeMain==UMIdedup::typeI::CR, true, umiCorrected[iG]);
if (pSolo.umiDedup.yes.Directional)
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + pSolo.umiDedup.countInd.Directional] =
umiArrayCorrect_Directional(nU0, umiArray, readInfo.size()>0 && pSolo.umiDedup.typeMain==UMIdedup::typeI::Directional, true, umiCorrected[iG], 0);
if (pSolo.umiDedup.yes.Directional_UMItools)
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + pSolo.umiDedup.countInd.Directional_UMItools] =
umiArrayCorrect_Directional(nU0, umiArray, readInfo.size()>0 && pSolo.umiDedup.typeMain==UMIdedup::typeI::Directional_UMItools, true, umiCorrected[iG], -1);
//this changes umiArray, so it should be last call
if (pSolo.umiDedup.yes.All)
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + pSolo.umiDedup.countInd.All] =
umiArrayCorrect_Graph(nU0, umiArray, readInfo.size()>0 && pSolo.umiDedup.typeMain==UMIdedup::typeI::All, true, umiCorrected[iG]);
};//if (nU0>0)
{//check any count>0 and finalize record for this gene
uint32 totcount=0;
for (uint32 ii=countCellGeneUMIindex[iCB+1]+1; ii<countCellGeneUMIindex[iCB+1]+countMatStride; ii++) {
totcount += countCellGeneUMI[ii];
};
if (totcount>0) {//at least one umiDedup type is non-0
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + 0] = gID[iG];
nGenePerCB[iCB]++;
nUMIperCB[iCB] += countCellGeneUMI[countCellGeneUMIindex[iCB+1] + pSolo.umiDedup.countInd.main];
countCellGeneUMIindex[iCB+1] = countCellGeneUMIindex[iCB+1] + countMatStride;//iCB+1 accumulates the index
};
};
if (readInfo.size()>0) {//record cb/umi for each read
for (uint32 iR=0; iR<gReadS[iG+1]-gReadS[iG]; iR+=rguStride) {//cycle over reads
uint64 iread1 = rGU1[iR+rguR];
readInfo[iread1].cb = indCB[iCB] ;
uint32 umi=rGU1[iR+rguU];
if (umiCorrected[iG].count(umi)>0)
umi=umiCorrected[iG][umi]; //correct UMI
readInfo[iread1].umi=umi;
};
};
};
if (pSolo.umiFiltering.MultiGeneUMI_CR) {
vector<uint32> geneCounts(nGenes,0);
vector<unordered_set<uintUMI>> geneUmiHash;
if (readInfo.size()>0)
geneUmiHash.resize(nGenes);
for (auto &iu: umiGeneHash) {//loop over UMIs for all genes
uint32 maxu=0, maxg=-1;
for (auto &ig : iu.second) {
if (ig.second>maxu) {
maxu=ig.second;
maxg=ig.first;
} else if (ig.second==maxu) {
maxg=-1;
};
};
if ( maxg+1==0 )
continue; //this umi is not counted for any gene, because two genes have the same read count for this UMI
for (auto &ig : umiGeneHash0[iu.first]) {//check that this umi/gene had also top count for uncorrected umis
if (ig.second>umiGeneHash0[iu.first][maxg]) {
maxg=-1;
break;
};
};
if ( maxg+1!=0 ) {//this UMI is counted
geneCounts[maxg]++;
if (readInfo.size()>0)
geneUmiHash[maxg].insert(iu.first);
};
};
for (uint32 ig=0; ig<nGenes; ig++) {
if (geneCounts[ig] == 0)
continue; //no counts for this gene
nGenePerCB[iCB]++;
nUMIperCB[iCB] += geneCounts[ig];
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + 0] = gID[ig];
countCellGeneUMI[countCellGeneUMIindex[iCB+1] + pSolo.umiDedup.countInd.CR] = geneCounts[ig];
countCellGeneUMIindex[iCB+1] = countCellGeneUMIindex[iCB+1] + countMatStride;//iCB+1 accumulates the index
};
if (readInfo.size()>0) {//record cb/umi for each read
for (uint32 iG=0; iG<nGenes; iG++) {//cycle over genes
uint32 *rGU1=rGU+gReadS[iG];
for (uint32 iR=0; iR<gReadS[iG+1]-gReadS[iG]; iR+=rguStride) {//cycle over reads
uint64 iread1 = rGU1[iR+rguR];
readInfo[iread1].cb = indCB[iCB] ;
uint32 umi=rGU1[iR+rguU];
if (umiCorrected[iG].count(umi)>0)
umi=umiCorrected[iG][umi]; //correct UMI
//cout << iG << "-" << iR << " " <<flush ;
if (geneUmiHash[iG].count(umi)>0) {
readInfo[iread1].umi=umi;
} else {
readInfo[iread1].umi=(uintUMI) -1;
};
};
};
};
};
};
////////////////////////////////////////////////////////////////////////////////////////////////
uint32 SoloFeature::umiArrayCorrect_CR(const uint32 nU0, uintUMI *umiArr, const bool readInfoRec, const bool nUMIyes, unordered_map <uintUMI,uintUMI> &umiCorr)
{
qsort(umiArr, nU0, umiArrayStride*sizeof(uint32), funCompareSolo1);
for (uint64 iu=0; iu<nU0*umiArrayStride; iu+=umiArrayStride) {
umiArr[iu+2] = umiArr[iu+0]; //stores corrected UMI for 1MM_CR and 1MM_Directional
for (uint64 iuu=(nU0-1)*umiArrayStride; iuu>iu; iuu-=umiArrayStride) {
uint32 uuXor = umiArr[iu+0] ^ umiArr[iuu+0];
if ( (uuXor >> (__builtin_ctz(uuXor)/2)*2) <= 3 ) {//1MM
umiArr[iu+2]=umiArr[iuu+0];//replace iu with iuu
break;
};
};
};
if (readInfoRec) {//record corrections
for (uint64 iu=0; iu<nU0*umiArrayStride; iu+=umiArrayStride) {
if (umiArr[iu+0] != umiArr[iu+2])
umiCorr[umiArr[iu+0]]=umiArr[iu+2];
};
};
if (!nUMIyes) {
return 0;
} else {
unordered_set<uintUMI> umiC;
for (uint64 iu=0; iu<nU0*umiArrayStride; iu+=umiArrayStride) {
umiC.insert(umiArr[iu+2]);
};
return umiC.size();
};
};
/////////////////////////////////////////////////////////////////////////////////////////////////////////
uint32 SoloFeature::umiArrayCorrect_Directional(const uint32 nU0, uintUMI *umiArr, const bool readInfoRec, const bool nUMIyes, unordered_map <uintUMI,uintUMI> &umiCorr, const int32 dirCountAdd)
{
qsort(umiArr, nU0, umiArrayStride*sizeof(uint32), funCompareNumbersReverseShift<uint32, 1>);//TODO no need to sort by sequence here, only by count.
for (uint64 iu=0; iu<nU0*umiArrayStride; iu+=umiArrayStride)
umiArr[iu+2] = umiArr[iu+0]; //initialized - it will store corrected UMI for 1MM_CR and 1MM_Directional
uint32 nU1 = nU0;
for (uint64 iu=umiArrayStride; iu<nU0*umiArrayStride; iu+=umiArrayStride) {
for (uint64 iuu=0; iuu<iu; iuu+=umiArrayStride) {
uint32 uuXor = umiArr[iu+0] ^ umiArr[iuu+0];
if ( (uuXor >> (__builtin_ctz(uuXor)/2)*2) <= 3 && umiArr[iuu+1] >= (2*umiArr[iu+1]+dirCountAdd) ) {//1MM && directional condition
umiArr[iu+2]=umiArr[iuu+2];//replace iuu with iu-corrected
nU1--;
break;
};
};
};
if (readInfoRec) {//record corrections
for (uint64 iu=0; iu<nU0*umiArrayStride; iu+=umiArrayStride) {
if (umiArr[iu+0] != umiArr[iu+2])
umiCorr[umiArr[iu+0]]=umiArr[iu+2];
};
};
if (!nUMIyes) {
return 0;
} else {
unordered_set<uintUMI> umiC;
for (uint64 iu=0; iu<nU0*umiArrayStride; iu+=umiArrayStride) {
umiC.insert(umiArr[iu+2]);
};
if (umiC.size()!=nU1)
cout << nU1 <<" "<< umiC.size()<<endl;
return umiC.size();
};
};
|
