int main(int argc, char* argv[]) {
char cpsfilename[MAXBUFFLENGTH];
char chainfilename[MAXBUFFLENGTH];
char outfilename[MAXBUFFLENGTH]="";
int marginlength = 0;
int MAXREC;
char c;
char *pc;
int x;
char buff[MAXBUFFLENGTH+1];
char aux[MAXBUFFLENGTH+1];
long score;
int start1,end1,len1,start2,end2,len2;
char strand1, strand2, chr1[MAXBUFFLENGTH], chr2[MAXBUFFLENGTH];
int *size, *dq, *dt;
int a,b,k,i,j,s,m;
int *position;
char *strand;
int *ids;
int *idg;
char *type;
int chridx[MAXCHR+1];
int chroff[MAXCHR+1];
char resstr;
int rescrd;
if(argc==1) {
fprintf(stderr,"Finds matches of the given set of sites (CPS file) in the BLASTZ chain alignment (CHAIN file)\n");
fprintf(stderr,"Last update by (dp) on Sep 21, 2011\n");
fprintf(stderr,"Keys:\n -i CPS file (remember to sort by position in ascending order)\n -d CHAIN alignment file\n -o output file\n");
fprintf(stderr," -m margin length [0]\n -v suppress verbose output [NO]\n");
exit(1);
}
timestamp_set();
for(i=1;i<argc;i++) {
pc = argv[i];
if(*pc != '-') continue;
if(*(pc+1) == 'i') {
sscanf(argv[++i], "%s", &cpsfilename[0]);
}
if(*(pc+1) == 'd') {
sscanf(argv[++i], "%s", &chainfilename[0]);
}
if(*(pc+1) == 'o') {
sscanf(argv[++i], "%s", &outfilename[0]);
}
if(*(pc+1) == 'm') {
sscanf(argv[++i], "%i", &marginlength);
}
if(*(pc+1) == 'v') {
verbose=0;
}
}
if(outfilename[0]==0) {
fprintf(stderr,"No output file privided, exiting\n");
exit(1);
}
outfile = fopen(outfilename,"w");
if(outfile == NULL) {
fprintf(stderr,"Can't open output file, exiting\n");
exit(1);
}
for(i=0;i<MAXCHR;i++) {
chridx[i]=chroff[i]=0;
}
MAXREC = 0;
cpsfile= fopen(cpsfilename,"r");
if(cpsfile==NULL) {
fprintf(stderr,"Can't access CPS file. Exiting\n");
exit(1);
}
if(verbose) fprintf(stderr,"Reading CPS input pass 1");
while(!feof(cpsfile)) {
buff[0]=0;
fgets(buff,MAXBUFFLENGTH,cpsfile);
if(strlen(buff)<2) break;
sscanf(buff,"%s" , aux);
chridx[assign_code(aux)]++;
MAXREC++;
}
fclose(cpsfile);
for(s=i=0;i<MAXCHR;i++) {
x = chridx[i];
chridx[i] =s;
s+=x;
}
chridx[i] = s;
position = (int*) malloc(sizeof(int)*(s+4));
strand = (char*) malloc(sizeof(char)*(s+4));
type = (char*) malloc(sizeof(char)*(s+4));
ids = (int*) malloc(sizeof(int)*(s+4));
idg = (int*) malloc(sizeof(int)*(s+4));
if(position==NULL || strand==NULL || type==NULL || ids==NULL || idg==NULL) {
fprintf(stderr,"Not enough memory. Terminated\n");
exit(1);
}
cpsfile= fopen(cpsfilename,"r");
if(verbose) fprintf(stderr,", records = %i\nReading CPS input pass 2",MAXREC);
while(!feof(cpsfile)) {
buff[0]=0;
fgets(buff,MAXBUFFLENGTH,cpsfile);
if(strlen(buff)<2) break;
sscanf(buff,"%s" , aux);
i = assign_code(aux);
m = chridx[i]+chroff[i];
sscanf(buff,"%*s %i %c %i %i %c" , position+m, strand+m, idg+m, ids+m, type+m);
chroff[i]++;
}
fclose(cpsfile);
if(verbose) fprintf(stderr,"\nSorting segments");
/*
for(i=0;i<MAXCHR;i++) {
quickSort_ic(position,strand,chridx[i],chridx[i+1]-1);
}
*/
for(i=0;i<MAXCHR;i++) {
k=1;
while(k) {
k=0;
for(j=chridx[i];j<chridx[i+1]-1;j++) {
if(position[j]>position[j+1]) {
k=1;
swapi(position+j,position+j+1);
swapc(strand+j,strand+j+1);
swapc(type+j,type+j+1);
swapi(ids+j,ids+j+1);
swapi(idg+j,idg+j+1);
}
}
}
}
if(verbose) fprintf(stderr," done\nProcessing chains");
/**********************************************************************************************/
size = (int*) malloc(sizeof(int)*MAXALN);
dq = (int*) malloc(sizeof(int)*MAXALN);
dt = (int*) malloc(sizeof(int)*MAXALN);
if(size ==0 || dq ==0 || dt==0) {
fprintf(stderr,"Not enough memory for such long chains. Terminated\n");
exit(1);
}
/**********************************************************************************************/
chainfile = fopen(chainfilename,"r");
while(!feof(chainfile)) {
buff[0]=0;
fgets(buff,MAXBUFFLENGTH,chainfile);
if(strlen(buff)<2) break;
buff[5]=0;
if(strcmp(buff,"chain")==0) {
sscanf(buff+6,"%li %s %i %c %i %i %s %i %c %i %i",&score, &chr1[0], &len1, &strand1, &start1, &end1, &chr2[0], &len2, &strand2, &start2, &end2);
k=0;
while(!feof(chainfile)) {
buff[0]=0;
fgets(buff,MAXBUFFLENGTH,chainfile);
if(strlen(buff)<2) break;
sscanf(buff,"%i %i %i",&size[k],&dt[k],&dq[k]);
k++;
if(k>MAXALN) {
fprintf(stderr,"Chain length exceeded. Terminating");
exit(1);
}
}
x = get_chr_code(chr1);
if(x<0) continue;
a=start1;b=start2;
j=0;
for(i=chridx[x];i<chridx[x+1] && position[i]<start1;i++);
for(;i<chridx[x+1]&& position[i]<end1;i++) {
while(position[i]>a+size[j]+dt[j] && j<k){
a+=size[j]+dt[j];
b+=size[j]+dq[j];
j++;
}
if(j>=k) break;
if(position[i]-a > marginlength && a+size[j]-position[i] >= marginlength) {
if(strand1==strand2) {
resstr = strand[i];
rescrd = position[i] - a + b;
}
else {
resstr = (strand[i]=='+') ? '-' : '+';
rescrd = len2 - (position[i] - a + b - 1) ;
}
fprintf(outfile,"%s\t%i\t%c\t%s\t%i\t%c\t%i\t%i\t%c\t%li\n",chr1, position[i], strand[i],chr2,rescrd, resstr, idg[i], ids[i],type[i],score);
}
}
}
}
if(verbose) fprintf(stderr," done\n");
fclose(chainfile);
fclose(outfile);
timestamp_report();
exit(0);
}
int main(int argc, char* argv[]) {
char out_file_name[MAXBUFFLENGTH]="";
char input_file_name[MAXBUFFLENGTH]="";
char idx_file_name[MAXBUFFLENGTH];
char dbx_file_name[MAXBUFFLENGTH];
long offset;
long seqlen;
long intronic_window = 150;
long exonic_window = 0;
char filename[MAXBUFFLENGTH];
char name[MAXBUFFLENGTH];
char chr_name[MAXBUFFLENGTH];
char buff[MAXBUFFLENGTH];
char longbuff[MAXLONGBUFFLENGTH];
char longbuffm[MAXLONGBUFFLENGTH];
FILE *idx_file;
FILE *dbx_file;
FILE *input_file;
FILE *outfile;
int b,i,j,q,k,n,a,m,s;
char c;
long position,y;
int strand;
long** pos;
int** str;
char** typ;
char*** ids;
long p,l;
int record_count[MAXCHR];
int record_idx[MAXCHR];
int cis = 0;
int coord = 0;
int all=0;
int warnings = 0;
char format[][64] = {"%*s %*i %*i %s %li %i %*s %s %c", "%s %li %i %*s %s %c"};
if(argc==1) {
fprintf(stderr,"This routine get sequence segments from a custom compressed FASTA repository (see transf)\n");
fprintf(stderr,"Last update by Dmitri Pervouchine ([email protected]) on Mar 22, 2013\n");
fprintf(stderr," -in <aln_file>\n -dbx <database_file>\n -idx <index_file>\n -out <output_file>\n");
fprintf(stderr," -we <exonic_window> [default=%i]\n -wi <intronic_window> [default=%i]\n -cis [use colunms 1-3] [default=%i]\n", exonic_window, intronic_window, cis);
fprintf(stderr," -quiet <suppress verbose output> [default=no]\n -all <include all sites>\n -coord <offset for 3'-sites> [default=%i]\n",coord);
exit(1);
}
timestamp_set();
for(i=1;i<argc;i++) {
if(strcmp(argv[i],"-in")==0) {
sscanf(argv[++i], "%s", &input_file_name[0]);
}
if(strcmp(argv[i],"-dbx")==0) {
sscanf(argv[++i], "%s", &dbx_file_name[0]);
}
if(strcmp(argv[i],"-idx")==0) {
sscanf(argv[++i], "%s", &idx_file_name[0]);
}
if(strcmp(argv[i],"-out")==0) {
sscanf(argv[++i], "%s", &out_file_name[0]);
}
if(strcmp(argv[i],"-we")==0) {
sscanf(argv[++i], "%li", &exonic_window);
}
if(strcmp(argv[i],"-wi")==0) {
sscanf(argv[++i], "%li", &intronic_window);
}
if(strcmp(argv[i],"-coord")==0) {
sscanf(argv[++i], "%i", &coord);
}
if(strcmp(argv[i],"-quiet")==0) {
verbose = 0;
}
if(strcmp(argv[i],"-cis")==0) {
cis = 1;
}
if(strcmp(argv[i],"-all")==0) {
all = 1;
}
}
if(out_file_name[0]==0) {
fprintf(stderr,"[WARNING: output file not specified, redirect to stdout]\n");
outfile = stdout;
}
else {
outfile = fopen(out_file_name,"w");
if(outfile == NULL) {
fprintf(stderr,"[ERROR: output file %s cannot be opened for writing, exiting]\n", out_file_name);
exit(1);
}
if(verbose) fprintf(stderr,"[>%s]\n",out_file_name);
}
input_file = fopen(input_file_name,"r");
if(input_file == NULL) {
fprintf(stderr,"[ERROR: cannot access %s, exiting]\n", input_file_name);
exit(1);
}
if(verbose) fprintf(stderr,"[<%s, pass 1",input_file_name);
while(fgets(buff,MAXBUFFLENGTH,input_file)) {
if(strlen(buff)<2) break;
sscanf(buff, format[cis], &chr_name[0], &position, &strand, &name[0], &c);
n = assign_code(chr_name);
record_count[n]++;
}
if(verbose) fprintf(stderr,"]\n");
pos = (long**) malloc(sizeof(long*)*(N_CHR_NAMES+1));
str = (int**) malloc(sizeof(int*)*(N_CHR_NAMES+1));
typ = (char**) malloc(sizeof(char*)*(N_CHR_NAMES+1));
ids = (char***)malloc(sizeof(char**)*(N_CHR_NAMES+1));
if(pos==NULL || str==NULL || typ==NULL || ids==NULL) {
fprintf(stderr,"[ERROR: failed to create index tables, exiting]\n");
exit(1);
}
for(i=0;i<N_CHR_NAMES;i++) {
if(record_count[i]>0) {
pos[i] = (long*) malloc(sizeof(long)*(record_count[i]+1));
str[i] = (int*) malloc(sizeof(int)*(record_count[i]+1));
typ[i] = (char*) malloc(sizeof(char)*(record_count[i]+1));
ids[i] = (char**) malloc(sizeof(char*)*(record_count[i]+1));
if(pos[i]==NULL || str[i]==NULL || typ[i]==NULL || ids[i]==NULL) {
fprintf(stderr,"[ERROR: failed to create index tables, exiting]\n");
exit(1);
}
record_idx[i]=0;
}
}
if(verbose) fprintf(stderr,"[<%s, pass 2",input_file_name);
fseek(input_file, 0, SEEK_SET);
while(fgets(buff,MAXBUFFLENGTH,input_file)) {
if(strlen(buff)<2) break;
sscanf(buff, format[cis], &chr_name[0], &position, &strand, &name[0], &c);
i = get_chr_code(chr_name);
j = record_idx[i];
pos[i][j] = position + (c=='D' ? coord*strand : 0);
str[i][j] = strand;
typ[i][j] = c;
ids[i][j] = (char*) malloc(sizeof(char)*(strlen(name)+1));
strcpy(ids[i][j],name);
record_idx[i]++;
}
if(verbose) fprintf(stderr,"]\n");
if(verbose) fprintf(stderr,"[<%s,%s",idx_file_name,dbx_file_name);
idx_file = fopen(idx_file_name,"r");
dbx_file = fopen(dbx_file_name,"r");
if(idx_file == NULL || dbx_file == NULL) {
fprintf(stderr,"[ERROR: cannot access %s or %s, exiting]\n", idx_file_name, dbx_file_name);
exit(1);
}
offset = 0;
while(fgets(buff,MAXBUFFLENGTH,idx_file)) {
if(strlen(buff)<2) break;
sscanf(buff,"%s" , &name[0]);
while(fgets(buff,MAXBUFFLENGTH,idx_file)) {
if(strlen(buff)<2) break;
sscanf(buff,"%s %li" , &chr_name[0], &seqlen);
i = get_chr_code(chr_name);
for(k=0;k<record_count[i];k++) {
if(pos[i][k]>seqlen) {
warnings++;
continue;
}
l = exonic_window + intronic_window;
if(typ[i][k]=='D' || typ[i][k]=='A' || all) {
if(str[i][k]>0) {
p = pos[i][k] - 1 - (typ[i][k]=='D' ? exonic_window : intronic_window);
}
else {
p = pos[i][k] - (typ[i][k]=='D' ? intronic_window : exonic_window);
}
fget_segment(longbuff, dbx_file, offset, p, l);
if(str[i][k]<0) {
rev1(longbuff);
}
if(is_all_n(longbuff)) continue;
fprintf(outfile,"%s\t%s\t%li\t%li\t%i\t%li\t%s\t%c\n",ids[i][k], chr_name, (str[i][k]>0 ? p + 1 : seqlen - (p + l)), l, str[i][k], seqlen, longbuff,typ[i][k]);
}
}
offset+= (seqlen % 8 == 0) ? seqlen/8 : (seqlen/8 + 1);
}
}
fclose(outfile);
fclose(idx_file);
fclose(dbx_file);
if(verbose) fprintf(stderr,"]\n");
if(verbose && warnings>0) fprintf(stderr,"[WARNING: %i windows were out of range, they were ignored]\n", warnings);
timestamp_report();
exit(0);
}
int main(int argc, char* argv[]) {
char cps_file_name[MAXBUFFLENGTH];
char chain_file_name[MAXBUFFLENGTH];
char out_file_name[MAXBUFFLENGTH]="";
int marginlength = 0;
char buff[MAXBUFFLENGTH+1];
char aux[MAXBUFFLENGTH+1];
long score;
int start1,end1,len1,start2,end2,len2;
char strand1, strand2, chr1[MAXBUFFLENGTH], chr2[MAXBUFFLENGTH];
int *size, *dq, *dt;
int a,b,k,i,j,s,m,x;
int *position;
int *strand;
int *idg;
int *ids;
char *type;
char c;
int chridx[MAXCHR+1];
int chroff[MAXCHR+1];
char resstr;
int rescrd;
long chain_id;
if(argc==1) {
fprintf(stderr,"This utility does liftOver of coordinates (cps) by using chain alignment\n");
fprintf(stderr,"Gene information is included in the output\n");
fprintf(stderr,"Last update by Dmitri Pervouchine ([email protected]) on Mar 22, 2013\n");
fprintf(stderr,"Usage: %s -in <cps_file> -chain <chain_alignment_file> [-margin <length>] [-quiet]\n", argv[0]);
fprintf(stderr," -in cps6, i.e. chr1/position1/strand1/gene/site/type tab-delimited file, strand is +/-\n");
fprintf(stderr," -chain UCSC chain alignment file, species1=>2\n");
fprintf(stderr," -out <output_file> [default=stdout]\n");
fprintf(stderr," -margin margin length [default=0]\n -quiet suppress verbose output [default=NO]\n");
fprintf(stderr,"NOTE: Input has to be sorted by position!\n");
fprintf(stderr,"Output format cps3+cps6: chr1/position1/strand1/chr2/position2/strand2/gene/site/type/score\n");
exit(1);
}
timestamp_set();
for(i=1;i<argc;i++) {
if(strcmp(argv[i],"-in")==0) {
sscanf(argv[++i], "%s", &cps_file_name[0]);
}
if(strcmp(argv[i],"-chain")==0) {
sscanf(argv[++i], "%s", &chain_file_name[0]);
}
if(strcmp(argv[i],"-out")==0) {
sscanf(argv[++i], "%s", &out_file_name[0]);
}
if(strcmp(argv[i],"-margin")==0) {
sscanf(argv[++i], "%i", &marginlength);
}
if(strcmp(argv[i],"-quiet")==0) {
verbose=0;
}
}
if(out_file_name[0]==0) {
fprintf(stderr,"[WARNING: output file not specified, redirect to stdout]\n");
out_file = stdout;
}
else {
out_file = fopen(out_file_name,"w");
if(out_file == NULL) {
fprintf(stderr,"[ERROR: output file %s cannot be opened for writing, exiting]\n", out_file_name);
exit(1);
}
if(verbose) fprintf(stderr,"[>%s]\n",out_file_name);
}
cps_file= fopen(cps_file_name,"r");
if(cps_file==NULL) {
fprintf(stderr,"[ERROR: cannot access %s, exiting]\n", cps_file_name);
exit(1);
}
for(i=0;i<MAXCHR;i++) chridx[i] = chroff[i] = 0;
if(verbose) fprintf(stderr,"[<%s, pass 1",cps_file_name);
while(fgets(buff,MAXBUFFLENGTH,cps_file)) {
if(strlen(buff)<2) break;
sscanf(buff,"%s" , aux);
chridx[assign_code(aux)]++;
}
if(verbose) fprintf(stderr,"]\n");
for(s=i=0;i<MAXCHR;i++) {
x = chridx[i];
chridx[i] =s;
s+=x;
}
chridx[i] = s;
position = (int*) malloc(sizeof(int)*(s + ARRAY_MARGIN));
strand = (int*) malloc(sizeof(int)*(s + ARRAY_MARGIN));
ids = (int*) malloc(sizeof(int)*(s + ARRAY_MARGIN));
idg = (int*) malloc(sizeof(int)*(s + ARRAY_MARGIN));
type = (char*) malloc(sizeof(char)*(s + ARRAY_MARGIN));
if(position==NULL || strand==NULL || type==NULL || ids==NULL || idg==NULL) {
fprintf(stderr,"[ERROR: failed to create index tables, exiting]\n");
exit(1);
}
fseek (cps_file, 0, SEEK_SET);
if(verbose) fprintf(stderr,"[<%s, pass 2", cps_file_name);
while(fgets(buff,MAXBUFFLENGTH,cps_file)) {
if(strlen(buff)<2) break;
sscanf(buff,"%s" , aux);
i = assign_code(aux);
m = chridx[i]+chroff[i];
sscanf(buff,"%*s %i %c %i %i %c" , &position[m], &c,&idg[m],&ids[m],&type[m]);
strand[m] = strand_c2i(c);
chroff[i]++;
}
fclose(cps_file);
if(verbose) fprintf(stderr,"]\n");
if(verbose) fprintf(stderr,"[Sort by position (if not done before)");
for(i=0;i<MAXCHR;i++) {
k=1;
while(k) {
k=0;
for(j=chridx[i];j<chridx[i+1]-1;j++) {
if(position[j]>position[j+1]) {
k=1;
swapi(position+j,position+j+1);
swapi(strand+j,strand+j+1);
swapi(idg+j,idg+j+1);
swapi(ids+j,ids+j+1);
swapc(type+j,type+j+1);
}
}
}
}
if(verbose) fprintf(stderr,"]\n");
/**********************************************************************************************/
size = (int*) malloc(sizeof(int)*(MAXALN + ARRAY_MARGIN));
dq = (int*) malloc(sizeof(int)*(MAXALN + ARRAY_MARGIN));
dt = (int*) malloc(sizeof(int)*(MAXALN + ARRAY_MARGIN));
if(size ==0 || dq ==0 || dt==0) {
fprintf(stderr,"[ERROR: not enough memory for chains, exiting]\n");
exit(1);
}
/**********************************************************************************************/
chain_file = fopen(chain_file_name,"r");
if(chain_file==NULL) {
fprintf(stderr,"[ERROR: cannot access %s, exiting]\n", chain_file_name);
exit(1);
}
fseek(chain_file, 0, SEEK_END);
unsigned int last_pos = ftell(chain_file);
fseek(chain_file, 0, SEEK_SET);
while(fgets(buff,MAXBUFFLENGTH,chain_file)) {
if(strlen(buff)<2) break;
buff[5]=0;
if(strcmp(buff,"chain")==0) {
sscanf(buff+6,"%li %s %i %c %i %i %s %i %c %i %i %li",&score, &chr1[0], &len1, &strand1, &start1, &end1, &chr2[0], &len2, &strand2, &start2, &end2, &chain_id);
k=0;
while(fgets(buff,MAXBUFFLENGTH,chain_file)) {
if(strlen(buff)<2) break;
progressbar(ftell(chain_file), last_pos-1, (char*)"Processing ", verbose);
sscanf(buff,"%i %i %i",size + k, dt + k, dq + k);
k++;
if(k>=MAXALN) {
fprintf(stderr,"[ERROR: chain too long, exiting]\n");
exit(1);
}
}
x = get_chr_code(chr1);
if(x<0) continue;
a=start1;b=start2;
j=0;
for(i=chridx[x];i<chridx[x+1] && position[i]<start1;i++);
for(;i<chridx[x+1]&& position[i]<end1;i++) {
while(position[i]>a+size[j]+dt[j] && j<k){
a+=size[j]+dt[j];
b+=size[j]+dq[j];
j++;
}
if(j>=k) break;
if(position[i]-a > marginlength && a+size[j]-position[i] >= marginlength) {
if(strand1==strand2) {
resstr = strand[i];
rescrd = position[i] - a + b;
}
else {
resstr = -strand[i];
rescrd = len2 - (position[i] - a + b - 1) ;
}
fprintf(out_file,"%s\t%i\t%c\t%s\t%i\t%c\t%i\t%i\t%c\t%li\n",chr1, position[i], strand_i2c(strand[i]), chr2, rescrd, strand_i2c(resstr), idg[i], ids[i],type[i],score);
}
}
}
}
fclose(chain_file);
fclose(out_file);
timestamp_report();
free(size);
free(dq);
free(dt);
free(position);
free(strand);
exit(0);
}
int main(int argc, char* argv[]) {
char alnfilename[MAXBUFFLENGTH];
char cpsfilename[MAXBUFFLENGTH];
char outfilename[MAXBUFFLENGTH]="";
char c;
char buff[MAXBUFFLENGTH];
char aux[MAXBUFFLENGTH];
int start1,end1,len1,start2,end2,len2;
char strand1, strand2;
int a,b,k,l,i,j,s,m,q;
int d,dmin,lmin,score_max;
int x,y;
int qbest, kbest, jbest;
int kprev;
int *gene_idx; //index
int *gene_off; //offset
int *gene_site; //site number
char *gene_styp; //site type
int *gene_pos; //site position
int *gene_chr; //chromosome
char *gene_str; //strand
int max_genes;
int max_sites;
int *site_idx; //index
int *site_off; //offset
int *site_chr; //matching chromosome
int *site_pos; //matching pos
int *site_str; //matching strand
int *site_score; //--- optimal score
int *site_lbest; //--- where it came from
int *site_qbest; //--- where it came from
int specific_site=0;
double dthreshold = 0.50;
int dlimit = 5000;
int max_depth = 4;
int pos, strand;
if(argc==1) {
fprintf(stderr,"Select best unique maping from the ALN file created by map_single\n");
fprintf(stderr,"Last updated by Dmitri Pervouchine ([email protected]) on Jan 28, 2013\n");
fprintf(stderr,"Keys:\n -in <cps file>\n -aln <aln file>\n -out <output file>\n");
//fprintf(stderr," -l length difference limit [%i]\n -t percentage difference threshold [%1.2lf] (ONE OR THE OTHER THRESHOLD IS USED)\n -h max_depth [%i]\n",dlimit, dthreshold,max_depth);
//fprintf(stderr," -v suppress verbose output [NO]\n");
exit(1);
}
timestamp_set();
for(i=1;i<argc;i++) {
if(strcmp(argv[i],"-in")==0) {
sscanf(argv[++i], "%s", &cpsfilename[0]);
}
if(strcmp(argv[i],"-aln")==0) {
sscanf(argv[++i], "%s", &alnfilename[0]);
}
if(strcmp(argv[i],"-out")==0) {
sscanf(argv[++i], "%s", &outfilename[0]);
}
if(strcmp(argv[i],"-lendiff")==0) {
sscanf(argv[++i], "%i", &dlimit);
}
if(strcmp(argv[i],"-threshold")==0) {
sscanf(argv[++i], "%lf", &dthreshold);
}
if(strcmp(argv[i],"-maxdepth")==0) {
sscanf(argv[++i], "%i", &max_depth);
}
if(strcmp(argv[i],"-quiet")==0) {
verbose=0;
}
if(strcmp(argv[i],"-s")==0) {
sscanf(argv[++i], "%i", &specific_site);
}
}
if(outfilename[0]==0) {
fprintf(stderr,"[WARNING: output file not specified, redirect to stdout]\n");
outfile = stdout;
}
else {
outfile = fopen(outfilename,"w");
if(outfile == NULL) {
fprintf(stderr,"[ERROR: output file (%s) cannot be opened, exiting]\n", outfilename);
exit(1);
}
}
/*******************************************************************************************************/
cpsfile= fopen(cpsfilename,"r");
if(cpsfile==NULL) {
fprintf(stderr,"Can't access CPS file. Exiting\n");
exit(1);
}
if(verbose) fprintf(stderr,"[Reading CPS, pass 0");
max_sites = max_genes = 0;
while(fgets(buff,MAXBUFFLENGTH,cpsfile)) {
if(strlen(buff)<2) break;
sscanf(buff,"%*s %*i %*i %i %i", &i, &j);
if(i>max_genes) max_genes = i;
if(j>max_sites) max_sites = j;
}
max_genes++;
max_sites++;
gene_idx = (int*) malloc(sizeof(int)*(max_genes+1));
gene_off = (int*) malloc(sizeof(int)*(max_genes+1));
gene_chr = (int*) malloc(sizeof(int)*(max_genes+1));
gene_str = (char*) malloc(sizeof(char)*(max_genes+1));
for(i=0;i<max_genes;i++) gene_idx[i]=gene_off[i]=0;
if(verbose) fprintf(stderr,", max_genes = %i, max_sites = %i]\n", max_genes, max_sites);
if(verbose) fprintf(stderr,"[Reading CPS, pass 1");
fseek (cpsfile, 0, SEEK_SET);
while(fgets(buff,MAXBUFFLENGTH,cpsfile)) {
if(strlen(buff)<2) break;
sscanf(buff,"%s %*i %i %i %*i" , aux, &strand, &i);
gene_idx[i]++;
gene_chr[i] = assign_code(aux);
gene_str[i] = strand;
}
for(s=i=0;i<max_genes;i++) {
x = gene_idx[i];
gene_idx[i] =s;
s+=x;
}
gene_idx[i] = s;
gene_site = (int*) malloc(sizeof(int)*(s+1));
gene_styp = (char*) malloc(sizeof(char)*(s+1));
gene_pos = (int*) malloc(sizeof(int)*(s+1));
if(verbose) fprintf(stderr,", records = %i]\n", s);
if(verbose) fprintf(stderr,"[Reading CPS, pass 2");
fseek (cpsfile, 0, SEEK_SET);
while(fgets(buff,MAXBUFFLENGTH,cpsfile)) {
if(strlen(buff)<2) break;
sscanf(buff,"%*s %i %*i %i %i %c" , &x, &i, &j, &c);
gene_site[gene_idx[i]+gene_off[i]]=j;
gene_styp[gene_idx[i]+gene_off[i]]=c;
gene_pos[gene_idx[i]+gene_off[i]]=x;
gene_off[i]++;
}
fclose(cpsfile);
if(verbose) fprintf(stderr,"]\n");
/**********************************************************************************************/
alnfile = fopen(alnfilename,"r");
if(alnfile == NULL) {
fprintf(stderr, "Cant open alignment file, exiting\n");
exit(1);
}
site_idx = (int*) malloc(sizeof(int)*(max_sites+1));
site_off = (int*) malloc(sizeof(int)*(max_sites+1));
for(i=0;i<max_sites;i++) {
site_idx[i]=site_off[i]=0;
}
if(verbose) fprintf(stderr,"[Reading alignment file, pass 1");
while(fgets(buff,MAXBUFFLENGTH,alnfile)) {
if(strlen(buff)<2) break;
sscanf(buff,"%*s %*i %*i %*s %*i %*i %*i %i %*c" , &i);
site_idx[i]++;
}
for(s=i=0;i<max_sites;i++) {
x = site_idx[i];
site_idx[i] =s;
s+=x;
}
site_idx[i] = s;
site_chr = (int*) malloc(sizeof(int)*(s+1));
site_pos = (int*) malloc(sizeof(int)*(s+1));
site_str = (int*) malloc(sizeof(int)*(s+1));
site_score = (int*) malloc(sizeof(int)*(s+1));
site_lbest = (int*) malloc(sizeof(int)*(s+1));
site_qbest = (int*) malloc(sizeof(int)*(s+1));
if(verbose) fprintf(stderr,", records = %i]\n",s);
if(verbose) fprintf(stderr,"[Reading alignment file, pass 2");
fseek (alnfile, 0, SEEK_SET);
while(fgets(buff,MAXBUFFLENGTH,alnfile)) {
if(strlen(buff)<2) break;
sscanf(buff,"%*s %*i %*i %s %i %i %*i %i %*c" , &aux, &pos, &strand, &i);
site_chr[site_idx[i]+site_off[i]] = assign_code(aux);
site_pos[site_idx[i]+site_off[i]] = pos;
site_str[site_idx[i]+site_off[i]] = strand;
site_score[site_idx[i]+site_off[i]] = 0;
site_lbest[site_idx[i]+site_off[i]] = -1;
site_qbest[site_idx[i]+site_off[i]] = -1;
site_off[i]++;
}
fclose(alnfile);
if(verbose) fprintf(stderr,"]\n");
for(i=0;i<max_genes;i++) {
progressbar(i,max_genes-1, (char*)"Processing");
score_max=0;
kbest = -1;
for(j=gene_idx[i];j<gene_idx[i+1];j++) {
x = gene_site[j];
for(k=site_idx[x];k<site_idx[x+1];k++) {
site_score[k] = 0;
site_lbest[k] = site_qbest[k] = -1;
}
for(q=1;q<=max_depth;q++) {
if(j-q>=gene_idx[i]) {
y = gene_site[j-q];
a = abs(gene_pos[j]-gene_pos[j-q]);
for(k=site_idx[x];k<site_idx[x+1];k++) {
dmin = INFTY;
lmin = -1;
for(l=site_idx[y];l<site_idx[y+1];l++) {
if(site_chr[k] == site_chr[l] && site_str[k] == site_str[l]) {
b = abs(site_pos[k]-site_pos[l]);
d = abs(b-a);
if(d<dmin) {
dmin = d;
lmin = l;
}
if(x==specific_site) fprintf(stderr,"[prev=%i curr=%i pos_p=%i pos_c=%i d=%i]\n",y,x,site_pos[l],site_pos[k],d);
}
}
m = (lmin>=0 && (((double)dmin/a)<dthreshold || dmin<dlimit)) ? site_score[lmin] + a : 0;
if(m>site_score[k]) {
site_score[k] = m;
site_lbest[k] = lmin;
site_qbest[k] = q;
}
if(site_score[k]>score_max) {
score_max = site_score[k];
kbest = k; jbest = j;
}
if(x==specific_site) fprintf(stderr,"[curr=%i score=%i]\n",x,site_score[k]);
}
}
}
}
j = jbest;
k = kbest;
if(k>=0 && site_score[k]>0) {
fprintf(outfile,"%s\t%i\t%i\t%s\t%i\t%i\t",get_chr_name(gene_chr[i]),gene_pos[j],gene_str[i],get_chr_name(site_chr[k]),site_pos[k],site_str[k]);
fprintf(outfile,"%i\t%i\t%c\t%i\t%i\n",i,gene_site[j],gene_styp[j],site_pos[site_lbest[k]],0);
while(site_score[k]>0 && site_lbest[k]>=0 && site_qbest[k]>=0) {
kprev = k;
j = j - site_qbest[k];
k = site_lbest[k];
fprintf(outfile,"%s\t%i\t%i\t%s\t%i\t%i\t",get_chr_name(gene_chr[i]),gene_pos[j],gene_str[i],get_chr_name(site_chr[k]),site_pos[k],site_str[k]);
fprintf(outfile,"%i\t%i\t%c\t%i\t%i\n",i,gene_site[j],gene_styp[j],(site_lbest[k]>=0?site_pos[site_lbest[k]]:0),(kprev>=0?site_pos[kprev]:0));
}
}
}
timestamp_report();
exit(0);
}
int main(int argc, char* argv[]) {
char aln_file_name[MAXBUFFLENGTH];
char out_file_name[MAXBUFFLENGTH]="";
char buff[MAXBUFFLENGTH];
char chr1[MAXBUFFLENGTH];
char chr2[MAXBUFFLENGTH];
double dthreshold = 1.50;
int dlimit = 100000;
int max_depth = 4;
int** chr_t;
int** str_t;
int** pos_t;
int* chr_q;
int* str_q;
int* pos_q;
int **score;
int **jbest;
int **lbest;
int a,b,d,dmin,lmin,s;
char c,c1,c2;
int *count;
int *ptr;
int max_rec=0;
int pos1, pos2, str1, str2;
int i, j, k, l, n;
int s_max, k_max;
if(argc==1) {
fprintf(stderr,"This utility takes a non-unique mapping in cps3+cps3 format and does ad-hoc filtering of the projected coordinates by maximum synteny\n");
fprintf(stderr,"Last update by Dmitri Pervouchine ([email protected]) on Mar 26, 2013\n");
fprintf(stderr,"Usage: %s -in <aln_file> -out <output_file> [-maxdepth <int>] [-threshold <double>] [-lendiff <diff>] [-quiet]\n",argv[0]);
fprintf(stderr," -in cps3+cps3 file, remember to sort by position in ascending order\n");
fprintf(stderr," -out <output_file> [default=stdout]\n");
fprintf(stderr," -maxdepth <integer> how many preceding positions can be skipped [default=%i]\n", max_depth);
fprintf(stderr," -threshold <double> max change of segment length, in percent [default=%2.2lf]\n", dthreshold);
fprintf(stderr," -lendiff <integer>, [default=%i]\n",dlimit);
fprintf(stderr," -quiet suppress verbose output [default=NO]\n");
fprintf(stderr,"Note: the mapping [x,x+dx] -> [y,y+dy] is OK if |dy-dx|/dx<threshold OR |dy-dx|<dlimit\n");
exit(1);
}
timestamp_set();
for(i=1;i<argc;i++) {
if(strcmp(argv[i],"-in")==0) {
sscanf(argv[++i], "%s", &aln_file_name[0]);
}
if(strcmp(argv[i],"-out")==0) {
sscanf(argv[++i], "%s", &out_file_name[0]);
}
if(strcmp(argv[i],"-lendiff")==0) {
sscanf(argv[++i], "%i", &dlimit);
}
if(strcmp(argv[i],"-threshold")==0) {
sscanf(argv[++i], "%lf", &dthreshold);
}
if(strcmp(argv[i],"-maxdepth")==0) {
sscanf(argv[++i], "%i", &max_depth);
}
if(strcmp(argv[i],"-quiet")==0) {
verbose=0;
}
}
if(out_file_name[0]==0) {
fprintf(stderr,"[WARNING: output file not specified, redirect to stdout]\n");
out_file = stdout;
}
else {
out_file = fopen(out_file_name,"w");
if(out_file == NULL) {
fprintf(stderr,"[ERROR: output file (%s) cannot be opened, exiting]\n", out_file_name);
exit(1);
}
if(verbose) fprintf(stderr,"[>%s]\n",out_file_name);
}
/*******************************************************************************************************/
aln_file= fopen(aln_file_name,"r");
if(aln_file==NULL) {
fprintf(stderr,"[ERROR: cannot access %s, exiting]\n", aln_file_name);
exit(1);
}
if(verbose) fprintf(stderr,"[<%s, pass 1", aln_file_name);
while(fgets(buff, MAXBUFFLENGTH, aln_file)) {
if(strlen(buff)<2) break;
max_rec++;
}
if(verbose) fprintf(stderr,"]\n");
chr_q = (int*)malloc(sizeof(int)*(max_rec + ARRAY_MARGIN));
str_q = (int*)malloc(sizeof(int)*(max_rec + ARRAY_MARGIN));
pos_q = (int*)malloc(sizeof(int)*(max_rec + ARRAY_MARGIN));
chr_t = (int**)malloc(sizeof(int*)*(max_rec + ARRAY_MARGIN));
str_t = (int**)malloc(sizeof(int*)*(max_rec + ARRAY_MARGIN));
pos_t = (int**)malloc(sizeof(int*)*(max_rec + ARRAY_MARGIN));
score = (int**)malloc(sizeof(int*)*(max_rec + ARRAY_MARGIN));
lbest = (int**)malloc(sizeof(int*)*(max_rec + ARRAY_MARGIN));
jbest = (int**)malloc(sizeof(int*)*(max_rec + ARRAY_MARGIN));
count = (int*)malloc(sizeof(int)*(max_rec + ARRAY_MARGIN));
ptr = (int*)malloc(sizeof(int)*(max_rec + ARRAY_MARGIN));
if(chr_q == NULL || str_q == NULL || pos_q == NULL || count == NULL || ptr == NULL || chr_t == NULL || str_t == NULL || pos_t == NULL) {
fprintf(stderr,"[ERROR: not enough memory, exiting]\n");
exit(1);
}
for(i=0;i<max_rec;i++) count[i] = ptr[i] = 0;
if(verbose) fprintf(stderr,"[<%s, pass 2", aln_file_name);
fseek (aln_file, 0, SEEK_SET);
n=0;
while(fgets(buff, MAXBUFFLENGTH, aln_file)) {
if(strlen(buff)<2) break;
sscanf(buff,"%s %i %c" , &chr1[0], &pos1, &c);
str1 = strand_c2i(c);
if(assign_code(chr1) != chr_q[n] || str1 != str_q[n] || pos1 != pos_q[n]) {
n++;
chr_q[n] = assign_code(chr1);
str_q[n] = str1;
pos_q[n] = pos1;
}
count[n]++;
}
if(verbose) fprintf(stderr,"]\n");
for(i=1; i<=n; i++) {
chr_t[i] = (int*)malloc(sizeof(int)*(count[i] + ARRAY_MARGIN));
pos_t[i] = (int*)malloc(sizeof(int)*(count[i] + ARRAY_MARGIN));
str_t[i] = (int*)malloc(sizeof(int)*(count[i] + ARRAY_MARGIN));
score[i] = (int*)malloc(sizeof(int)*(count[i] + ARRAY_MARGIN));
lbest[i] = (int*)malloc(sizeof(int)*(count[i] + ARRAY_MARGIN));
jbest[i] = (int*)malloc(sizeof(int)*(count[i] + ARRAY_MARGIN));
if(chr_t[i] == NULL || str_t[i] == NULL || pos_t[i] == NULL) {
fprintf(stderr,"[ERROR: not enough memory, exiting]\n");
exit(1);
}
}
if(verbose) fprintf(stderr,"[<%s, pass 3", aln_file_name);
fseek (aln_file, 0, SEEK_SET);
n=0;
while(fgets(buff, MAXBUFFLENGTH, aln_file)) {
if(strlen(buff)<2) break;
sscanf(buff,"%s %i %c %s %i %c" , &chr1[0], &pos1, &c1, &chr2[0], &pos2, &c2);
str1 = strand_c2i(c1);
str2 = strand_c2i(c2);
if(assign_code(chr1) != chr_q[n] || str1 != str_q[n] || pos1 != pos_q[n]) n++;
chr_t[n][ptr[n]] = assign_code(chr2);
pos_t[n][ptr[n]] = pos2;
str_t[n][ptr[n]] = str2;
ptr[n]++;
}
if(verbose) fprintf(stderr,"]\n");
for(i=1; i<=n; i++) {
progressbar(i, n, (char*)"Filtering ", verbose);
for(k=0; k<count[i]; k++) {
score[i][k] = 0;
lbest[i][k] = -1;
jbest[i][k] = -1;
for(j=i-1; j>0 && i-j<=max_depth; j--) {
a = abs(pos_q[i] - pos_q[j]);
dmin = INFTY;
lmin = -1;
for(l=0; l<count[j]; l++) {
if(chr_t[i][k] == chr_t[j][l] && str_t[i][k]*str_t[j][l] == str_q[i]*str_q[j]) {
b = abs(pos_t[i][k] - pos_t[j][l]);
d = abs(b-a);
if(d<dmin) {
dmin = d;
lmin = l;
}
}
}
s = (lmin>=0 && (((double)dmin/a) < dthreshold || dmin < dlimit)) ? score[j][lmin] + a : 0;
if(s > score[i][k]) {
score[i][k] = s;
jbest[i][k] = j;
lbest[i][k] = lmin;
}
}
}
}
for(i=n;i>0;i--) {
s_max = 0;
k_max = -1;
for(k=0; k<count[i]; k++) {
if(score[i][k]>s_max) {
s_max = score[i][k];
k_max = k;
}
}
if(k_max>=0) {
k = k_max;
while(jbest[i][k]>=0 && lbest[i][k]>=0) {
fprintf(out_file,"%s\t%i\t%c\t", get_chr_name(chr_q[i]), pos_q[i], strand_i2c(str_q[i]));
fprintf(out_file,"%s\t%i\t%c\n", get_chr_name(chr_t[i][k]), pos_t[i][k], strand_i2c(str_t[i][k]));
j = jbest[i][k];
l = lbest[i][k];
i = j;
k = l;
}
}
}
timestamp_report();
exit(0);
}
