void readGL2(gzFile fp,size_t nSites,int nChr,Matrix<double> &ret){
// fprintf(stderr,"[%s] fname:%s nSites:%d nChr:%d\n",__FUNCTION__,fname,nSites,nChr);
ret.y=nChr+1;
char buf[LENS];
size_t i=0;
while(gzgets(fp,buf,LENS)){
//strange construct for strtok, because firstcall if with buf, rest with NULL
for(char *tok = strtok(buf,"\t\n ");tok!=NULL;tok=strtok(NULL,"\t\n ")){
if(tok[0]=='#')
continue;
gzFile gz = getGz(buf);
while(SIG_COND&&gzread(gz,ret.mat[i],sizeof(double)*(nChr+1))){
for(size_t j=0;j<nChr+1;j++)
ret.mat[i][j] = exp(ret.mat[i][j]);
i++;
}
fprintf(stderr,"Done reading file: \'%s\'\n",buf);
gzclose(gz);
ret.x=i;
}
}
if(SIG_COND==0)
exit(0);
}
gzFile openFileGz(const char* a,const char* b,const char *mode){
if(0)
fprintf(stderr,"[%s] %s%s\n",__FUNCTION__,a,b);
char *c = new char[strlen(a)+strlen(b)+1];
strcpy(c,a);
strncat(c,b,strlen(b));
// fprintf(stderr,"\t-> Dumping file: %s\n",c);
dumpedFiles.push_back(strdup(c));
gzFile fp = getGz(c,mode);
delete [] c;
return fp;
}
size_t calcNsites(const char *fname,int nChr){
// fprintf(stderr,"[%s] fname: %s nChr: %d isList:%d\n",__FUNCTION__,fname,nChr,isList);
if(isList==0)
return subfun(fname,nChr);
else{
gzFile gz=getGz(fname);
char buf[LENS];
size_t tot=0;
while(gzgets(gz,buf,LENS)){
// fprintf(stderr,"buf:%s\n",buf);
for(char *tok = strtok(buf,"\n\t ");tok!=NULL;tok=strtok(NULL,"\n\t ")){
if(tok[0]=='#')
continue;
size_t per_file = subfun(buf,nChr);
tot+=per_file;
}
}
gzclose(gz);
return tot;
}
}
int main(int argc,char**argv){
if(argc!=5){
fprintf(stderr,"\nProgram extract ranges from output of msToGlf\n./splitgl file.glf.gz nindTotal firstInd lastInd\n");
fprintf(stderr,"\nTo extract first the GLS for the first 10 samples in an glf.gz that contains 25 samples\n");
fprintf(stderr,"Examples\n\n\t1) Extract sample 1 to 12 from a glf.gz file containing 20samples:\n\t\t\t./splitgl raw.glf.gz 20 1 12\n");
fprintf(stderr,"\t2) Extract sample 1 to 12 from a glf.gz file containing 20samples:\n\t\t\t./splitgl raw.glf.gz 20 13 20\n");
return 0;
}
const char *fname = argv[1];
int tot = atoi( argv[2]);
int first = atoi( argv[3]);
int last = atoi( argv[4]);
first--;last--;
fprintf(stderr,"fname:%s tot:%d first:%d last:%d\n",fname,tot,first,last);
gzFile gz=getGz(fname,"r");
double *lk = malloc(sizeof(double)*10*tot);
int nsites =0;
while(1){
int br=gzread(gz,lk,10*sizeof(double)*tot);
nsites++;
if(br==0)
break;
assert(br==10*sizeof(double)*tot);
int beg=10*first;
int nDoubles=(last-first+1)*10;
// fprintf(stderr,"beg: %d nDoubles:%d\n",beg,nDoubles);return 0;
br=fwrite(lk+beg,sizeof(double),nDoubles,stdout);
if(br!=nDoubles){
fprintf(stderr,"Problem writing full chunk\n");
return 0;
}
}
fprintf(stderr,"nsites:%d processed\n",nsites);
free(lk);
gzclose(gz);
}
int main_2dsfs(int argc,char **argv){
if(argc==1){
fprintf(stderr,"./emOptim2 2dsfs pop1 pop2 nChr1 nChr2 [-start FNAME -P nThreds -tole tole -maxIter ] (only works if the two saf files covers the same region)\n");
return 0;
}
argv++;
argc--;
fname1 = *(argv++);
fname2 = *(argv++);
argc -=2;
chr1 = atoi(*(argv++));
chr2 = atoi(*(argv++));
argc -=2;
getArgs(argc,argv);
if(nSites==0){
if(fsize(fname1)+fsize(fname2)>getTotalSystemMemory())
fprintf(stderr,"Looks like you will allocate too much memory, consider starting the program with a lower -nSites argument\n");
//this doesnt make sense if ppl supply a filelist containing safs
nSites=calcNsites(fname1,chr1);
}
fprintf(stderr,"fname1:%sfname2:%s chr1:%d chr2:%d startsfs:%s nThreads=%d tole=%f maxIter=%d nSites:%lu\n",fname1,fname2,chr1,chr2,sfsfname,nThreads,tole,maxIter,nSites);
float bytes_req_megs = nSites*(sizeof(double)*(chr1+1) + sizeof(double)*(chr2+1)+2*sizeof(double*))/1024/1024;
float mem_avail_megs = getTotalSystemMemory()/1024/1024;//in percentile
// fprintf(stderr,"en:%zu to:%f\n",bytes_req_megs,mem_avail_megs);
fprintf(stderr,"The choice of -nSites will require atleast: %f megabyte memory, that is approx: %.2f%% of total memory\n",bytes_req_megs,bytes_req_megs*100/mem_avail_megs);
#if 0
//read in positions, not used, YET...
std::vector<int> p1 = getPosi(fname1);
std::vector<int> p2 = getPosi(fname2);
fprintf(stderr,"nSites in pop1: %zu nSites in pop2: %zu\n",p1.size(),p2.size());
#endif
if(nSites==0){
if(calcNsites(fname1,chr1)!=calcNsites(fname2,chr2)){
fprintf(stderr,"Problem with number of sites in file: %s and %s\n",fname1,fname2);
exit(0);
}
nSites=calcNsites(fname1,chr1);
}
gzFile gz1=getGz(fname1);
gzFile gz2=getGz(fname2);
dim=(chr1+1)*(chr2+1);
Matrix<double> GL1=alloc(nSites,chr1+1);
Matrix<double> GL2=alloc(nSites,chr2+1);
dim=GL1.y*GL2.y;
double *sfs = new double[dim];
while(1){
if(isList ==0){
readGL(gz1,nSites,chr1,GL1);
readGL(gz2,nSites,chr2,GL2);
}else{
readGL2(gz1,nSites,chr1,GL1);
readGL2(gz2,nSites,chr2,GL2);
}
assert(GL1.x==GL2.x);
if(GL1.x==0)
break;
if(sfsfname!=NULL){
readSFS(sfsfname,dim,sfs);
}else{
for(int i=0;i<dim;i++)
sfs[i] = (i+1)/((double)dim);
normalize(sfs,dim);
}
setThreadPars(&GL1,&GL2,sfs,nThreads);
if(calcLike==0){
if(SIG_COND)
em2(sfs,&GL1,&GL2,tole,maxIter);
}
double lik;
if(nThreads>1)
lik = lik1_master();
else
lik = lik1(sfs,&GL1,0,GL1.x);
fprintf(stderr,"likelihood: %f\n",lik);
#if 1
int inc=0;
for(int x=0;x<chr1+1;x++){
for(int y=0;y<chr2+1;y++)
fprintf(stdout,"%f ",log(sfs[inc++]));
fprintf(stdout,"\n");
}
#endif
if(isList==1)
break;
}
dalloc(GL1,nSites);
dalloc(GL2,nSites);
gzclose(gz1);
gzclose(gz2);
return 0;
}
int main_1dsfs(int argc,char **argv){
if(argc<2){
fprintf(stderr,"Must supply afile.saf and number of chromosomes\n");
return 0;
}
fname1 = *(argv++);
chr1 = atoi(*(argv++));
argc-=2;
getArgs(argc,argv);
dim=chr1+1;
//hook for new EJ banded version
if(isNewFormat(fname1))
return main_1dsfs_v2(fname1,chr1,nSites,nThreads,sfsfname,tole,maxIter);
if(nSites==0){//if no -nSites is specified
if(fsize(fname1)>getTotalSystemMemory())
fprintf(stderr,"Looks like you will allocate too much memory, consider starting the program with a lower -nSites argument\n");
//this doesnt make sense if ppl supply a filelist containing safs
nSites=calcNsites(fname1,chr1);
}
fprintf(stderr,"fname1:%s nChr:%d startsfs:%s nThreads:%d tole=%f maxIter=%d nSites=%lu\n",fname1,chr1,sfsfname,nThreads,tole,maxIter,nSites);
float bytes_req_megs = nSites*(sizeof(double)*(chr1+1)+sizeof(double*))/1024/1024;
float mem_avail_megs = getTotalSystemMemory()/1024/1024;//in percentile
// fprintf(stderr,"en:%zu to:%f\n",bytes_req_megs,mem_avail_megs);
fprintf(stderr,"The choice of -nSites will require atleast: %f megabyte memory, that is approx: %.2f%% of total memory\n",bytes_req_megs,bytes_req_megs*100/mem_avail_megs);
Matrix<double> GL1=alloc(nSites,dim);
gzFile gz1=getGz(fname1);
double *sfs=new double[dim];
while(1) {
if(isList==0)
readGL(gz1,nSites,chr1,GL1);
else
readGL2(gz1,nSites,chr1,GL1);
if(GL1.x==0)
break;
fprintf(stderr,"dim(GL1)=%zu,%zu\n",GL1.x,GL1.y);
if(sfsfname!=NULL){
readSFS(sfsfname,dim,sfs);
}else{
for(int i=0;i<dim;i++)
sfs[i] = (i+1)/((double)dim);
if(doBFGS){
double ts=1;
for(int i=0;i<dim-1;i++)
ts += 0.01/(1.0+i);
sfs[0]=1.0/ts;
for(int i=0;i<dim-1;i++)
sfs[i+1] = (0.01/(1.0+i))/ts;
}
normalize(sfs,dim);
}
// em2_smart(sfs2,pops,1e-6,1e3);
setThreadPars(&GL1,NULL,sfs,nThreads);
if(calcLike==0){
if(doBFGS==0)
em1(sfs,&GL1,tole,maxIter);
else
bfgs(sfs,&GL1);
}
double lik;
if(nThreads>1)
lik = lik1_master();
else
lik = lik1(sfs,&GL1,0,GL1.x);
fprintf(stderr,"likelihood: %f\n",lik);
#if 1
for(int x=0;x<dim;x++)
fprintf(stdout,"%f ",log(sfs[x]));
fprintf(stdout,"\n");
fflush(stdout);
#endif
if(isList==1)
break;
}
dalloc(GL1,nSites);
gzclose(gz1);
delete [] sfs;
return 0;
}
int main(int argc, char **argv){
fprintf(stderr,"NGSrelate buildtime: (%s:%s)\n",__DATE__,__TIME__);
if(argc==1){// if no arguments, print info on program
info();
return 0;
}
//below for catching ctrl+c, and dumping files
struct sigaction sa;
sigemptyset (&sa.sa_mask);
sa.sa_flags = 0;
sa.sa_handler = handler;
sigaction(SIGPIPE, &sa, 0);
sigaction(SIGINT, &sa, 0);
//initial values
char *bfile = NULL;
char *binfile = NULL;
char *outfiles = NULL;
double a,k0,k1,k2;
a=k0=k1=-1;
k2=0;
int calcA = 1;
//filter opts
double minMaf = 0.01;
char *freqfile =NULL;
// reading arguments
argv++;
while(*argv){
if(strcmp(*argv,"-beagle")==0 ) bfile=*++argv;
else if(strcmp(*argv,"-bin")==0 ) binfile=*++argv;
else if(strcmp(*argv,"-freqfile")==0 ) freqfile=*++argv;
else if(strcmp(*argv,"-outnames")==0 ) outfiles=*++argv;
else if(strcmp(*argv,"-minMaf")==0 ) minMaf = atoi(*++argv);
else if(strcmp(*argv,"-a")==0 ) a = atof(*++argv);
else if(strcmp(*argv,"-k0")==0 ) k0 = atof(*++argv);
else if(strcmp(*argv,"-k1")==0 ) k1 = atof(*++argv);
else if(strcmp(*argv,"-k2")==0 ) k2 = atof(*++argv);
else if(strcmp(*argv,"-calcA")==0 ) calcA = atoi(*++argv);
else{
fprintf(stderr,"Unknown arg:%s\n",*argv);
info();
return 0;
}
++argv;
}
if(bfile==NULL&&binfile==NULL){
fprintf(stderr,"Please supply input data file: -beagle OR -bin");
info();
}else if(bfile!=NULL&&binfile!=NULL){
fprintf(stderr,"Please supply input data file: -beagle OR -bin");
info();
}
if(outfiles==NULL){
if(bfile!=NULL)
outfiles=bfile;
else
outfiles = binfile;
fprintf(stderr,"Will use: %s as prefix for output\n",outfiles);
}
FILE *flog=openFile(outfiles,".log");
clock_t t=clock();//how long time does the run take
time_t t2=time(NULL);
std::vector<perChr> pd;
if(bfile!=NULL){
bgl d=readBeagle(bfile);
fprintf(stderr,"Input beaglefile has dim: nsites=%d nind=%d\n",d.nSites,d.nInd);
pd = makeDat(d);
gzFile dfile = openFileGz(outfiles,".bin.gz","wb");
for(uint i=0;i<pd.size();i++)
marshall_dump(dfile,pd[i]);
gzclose(dfile);
}else{
gzFile dfile = getGz(binfile,"rb");
pd = marshall_read(dfile);
gzclose(dfile);
}
//calculate frequencies
for(uint i=0;i<pd.size();i++)
setFreq(pd[i]);
// printStuff(pd);
//set seed
srand(seed);
//below is the main looping trhought the iterations.
FILE *fp =openFile(outfiles,".freq");
for(int i=0;i<pd[0].nSites;i++)
fprintf(fp,"%f ",exp(pd[0].freq[i]));
fclose(fp);
double *freq = NULL;
if(freqfile!=NULL){
freq = readDouble(freqfile,pd[0].nSites);
fprintf(stderr,"freq=%f\n",freq[0]);
}
para p;
p.pair[0] = 0;p.pair[1] = 1;
p.a=a;p.k0=k0;p.k1=k1;p.k2=k2;
fprintf(stderr,"pa=%f pk0=%f\n",p.a,p.k0);
hmm res = analysis(pd[0],freq,p,calcA);
gzFile bo = openFileGz(outfiles,".bres.gz","wb");
fdump(bo,res,pd[0].name);
//deallocate memory
for(int i=0;1&&i<dumpedFiles.size();i++){
fprintf(stderr,"dumpedfiles are: %s\n",dumpedFiles[i]);
free(dumpedFiles[i]);
}
fprintf(stderr, "\t[ALL done] cpu-time used = %.2f sec\n", (float)(clock() - t) / CLOCKS_PER_SEC);
fprintf(stderr, "\t[ALL done] walltime used = %.2f sec\n", (float)(time(NULL) - t2));
// print to log file
fprintf(flog, "\t[ALL done] cpu-time used = %.2f sec\n", (float)(clock() - t) / CLOCKS_PER_SEC);
fprintf(flog, "\t[ALL done] walltime used = %.2f sec\n", (float)(time(NULL) - t2));
fclose(flog);
gzclose(bo);
return 0;
}
