static int
translate_chromosomepos_universal (unsigned int *genomicstart, unsigned int *genomiclength,
char *chromosome, unsigned int left, unsigned int length,
IIT_T chromosome_iit) {
int rc = 1, index;
Interval_T interval;
#ifdef DEBUG
bool allocp;
#endif
if ((index = IIT_find_linear(chromosome_iit,chromosome)) >= 0) {
debug(printf("chromosome %s => index %d\n",chromosome,index));
interval = IIT_interval(chromosome_iit,index);
debug(printf(" => label %s with interval low %u\n",
IIT_label(chromosome_iit,index,&allocp),Interval_low(interval)));
*genomicstart = Interval_low(interval)+left;
if (*genomicstart < Interval_low(interval)) {
fprintf(stderr,"%u + %u = %u (exceeds a 32-bit unsigned int)\n",
Interval_low(interval),left,*genomicstart);
exit(9);
}
if (length == 0) {
*genomiclength = Interval_length(interval)-left;
} else {
*genomiclength = length;
}
rc = 0;
}
return rc;
}
static void
find_positions (bool *revcompp, Genomicpos_T *leftposition, Genomicpos_T *rightposition,
Genomicpos_T *startposition, Genomicpos_T *endposition, Genomicpos_T *truelength,
int *contigtype, char *accession, IIT_T contig_iit) {
int index;
Interval_T interval;
char firstchar;
if ((index = IIT_find_one(contig_iit,accession)) == -1) {
fprintf(stderr,"Can't find accession %s in contig IIT file\n",
accession);
exit(9);
} else {
interval = IIT_interval(contig_iit,index);
*leftposition = Interval_low(interval);
*rightposition = Interval_high(interval);
if (IIT_version(contig_iit) <= 1) {
firstchar = IIT_annotation_firstchar(contig_iit,index);
if (firstchar == '-') {
*revcompp = true;
*startposition = Interval_high(interval) + 1U;
*endposition = Interval_low(interval) + 1U;
} else {
*revcompp = false;
*startposition = Interval_low(interval);
*endposition = Interval_high(interval);
}
} else {
if (Interval_sign(interval) < 0) {
*revcompp = true;
*startposition = Interval_high(interval) + 1U;
*endposition = Interval_low(interval) + 1U;
} else {
*revcompp = false;
*startposition = Interval_low(interval);
*endposition = Interval_high(interval);
}
}
*truelength = Interval_length(interval);
*contigtype = Interval_type(interval);
debug(printf("revcompp = %d, leftposition = %d, rightposition = %d, startposition = %d, endposition = %d\n",
*revcompp,*leftposition,*rightposition,*startposition,*endposition));
return;
}
}
/* Need to store just the part of the query specified (e.g., 1..10) */
static double
compute_logtotal_tally (long int *total, int *n, Chrpos_T coordstart, Chrpos_T coordend,
int indexi, IIT_T iit) {
double logtotal = 0.0;
Interval_T interval;
char *annotation, *restofheader, *ptr;
bool allocp;
Chrpos_T chrpos, intervalend;
long int count;
annotation = IIT_annotation(&restofheader,iit,indexi,&allocp);
interval = IIT_interval(iit,indexi);
chrpos = Interval_low(interval);
intervalend = Interval_high(interval);
ptr = annotation;
while (chrpos < coordstart) {
if ((ptr = index(ptr,'\n')) == NULL) {
fprintf(stderr,"Premature end of tally from %u to %u\n",
Interval_low(interval),Interval_high(interval));
return logtotal;
} else {
ptr++;
}
chrpos++;
}
while (chrpos <= intervalend && chrpos <= coordend) {
count = 0;
ptr = get_total_tally(&count,ptr);
logtotal += log((double) count + 1.0);
*total += count;
*n += 1;
ptr++;
chrpos++;
}
if (allocp == true) {
FREE(restofheader);
}
return logtotal;
}
static int
process_snp_block (int *nwarnings, Positionsptr_T *offsets, Genomicpos_T *positions,
Interval_T *intervals, int nintervals,
Genomicpos_T chroffset, Genome_T genome,
UINT4 *snp_blocks, int divno, char *divstring, int intervali,
IIT_T snps_iit, IIT_T chromosome_iit, int index1part) {
int nerrors = 0;
bool *snpp;
char *refstring;
char *altstring;
#ifdef DEBUG1
char *nt;
#endif
int index, length;
int nsnps, stringi, starti, shift, i, k;
char *snptype, *label, refnt, altnt;
unsigned int ptr;
Genomicpos_T snpposition, startposition, endposition, first_snppos, last_snppos, position, chrpos;
Chrnum_T chrnum;
Interval_T interval;
int nunknowns;
bool badcharp, allocp;
Uintlist_T oligomers, newoligomers, p;
Storedoligomer_T oligo;
#ifdef WORDS_BIGENDIAN
UINT4 high, low, flags;
#endif
/* Subtract 1 because snps_iit is 1-based */
first_snppos = Interval_low(intervals[0]) - 1U;
last_snppos = Interval_low(intervals[nintervals-1]) - 1U;
debug1(
if (nintervals == 1) {
printf("Processing snp at chrpos %s:%u\n",divstring,first_snppos+1U);
} else {
printf("Processing block of %d snps from chrpos %s:%u to %u\n",
nintervals,divstring,first_snppos+1U,last_snppos+1U);
}
);
static int
translate_contig (unsigned int *genomicstart, unsigned int *genomiclength,
char *contig, unsigned int left, unsigned int length, IIT_T contig_iit) {
int rc = 1, index;
Interval_T interval;
if ((index = IIT_find_one(contig_iit,contig)) >= 0) {
interval = IIT_interval(contig_iit,index);
*genomicstart = Interval_low(interval)+left;
if (length == 0) {
*genomiclength = Interval_length(interval)-left;
} else {
*genomiclength = length;
}
rc = 0;
}
return rc;
}
/* Need to store just the part of the query specified (e.g., 1..10) */
static void
print_interval_runlength (Chrpos_T *lastcoord, char *divstring, Chrpos_T coordstart, Chrpos_T coordend,
int indexi, IIT_T iit, bool zeroesp) {
Interval_T interval;
char *label;
bool allocp;
Chrpos_T chrpos, intervalend;
int value;
label = IIT_label(iit,indexi,&allocp);
value = atoi(label);
interval = IIT_interval(iit,indexi);
chrpos = Interval_low(interval);
intervalend = Interval_high(interval);
if (zeroesp == true) {
while (*lastcoord < chrpos) {
printf("%s\t%u\t%d\n",divstring,*lastcoord,0);
(*lastcoord)++;
}
}
while (chrpos < coordstart) {
chrpos++;
}
while (chrpos <= intervalend && chrpos <= coordend) {
printf("%s\t%u\t%d\n",divstring,chrpos,value);
chrpos++;
}
*lastcoord = chrpos;
if (allocp == true) {
FREE(label);
}
return;
}
/* coordstart used only if centerp or tallyp is true */
static long int
print_interval (Chrpos_T *lastcoord, long int total,
char *divstring, Chrpos_T coordstart, Chrpos_T coordend,
int index, IIT_T iit, int ndivs, int fieldint) {
Interval_T interval;
char *label, *annotation, *restofheader;
bool allocp;
if (centerp == true) {
print_interval_centered(divstring,coordstart,index,iit,fieldint);
return 0;
} else if (tallyp == true) {
total += print_interval_tally(&(*lastcoord),divstring,coordstart,coordend,index,iit,zeroesp);
return total;
} else if (runlengthp == true) {
print_interval_runlength(&(*lastcoord),divstring,coordstart,coordend,index,iit,zeroesp);
return 0;
}
if (annotationonlyp == false) {
label = IIT_label(iit,index,&allocp);
printf(">%s ",label);
if (allocp == true) {
FREE(label);
}
if (ndivs > 1) {
if (divstring == NULL) {
/* For example, if interval was retrieved by label */
divstring = IIT_divstring_from_index(iit,index);
}
printf("%s:",divstring);
}
debug(printf("index is %d\n",index));
interval = IIT_interval(iit,index);
if (signedp == false) {
printf("%u..%u",Interval_low(interval),Interval_high(interval));
} else if (Interval_sign(interval) < 0) {
printf("%u..%u",Interval_high(interval),Interval_low(interval));
} else {
printf("%u..%u",Interval_low(interval),Interval_high(interval));
}
if (Interval_type(interval) > 0) {
printf(" %s",IIT_typestring(iit,Interval_type(interval)));
}
#if 0
/* Unnecessary because of "\n" after restofheader below */
if (IIT_version(iit) < 5) {
printf("\n");
}
#endif
}
if (fieldint < 0) {
annotation = IIT_annotation(&restofheader,iit,index,&allocp);
printf("%s\n",restofheader);
printf("%s",annotation);
if (allocp == true) {
FREE(restofheader);
}
} else {
annotation = IIT_annotation(&restofheader,iit,index,&allocp);
printf("%s\n",restofheader);
if (allocp == true) {
FREE(restofheader);
}
annotation = IIT_fieldvalue(iit,index,fieldint);
printf("%s\n",annotation);
FREE(annotation);
}
return 0;
}
/* Need to store just the part of the query specified (e.g., 1..10) */
static long int
print_interval_tally (Chrpos_T *lastcoord, char *divstring, Chrpos_T coordstart, Chrpos_T coordend,
int indexi, IIT_T iit, bool zeroesp) {
long int total = 0, subtotal;
Interval_T interval;
char *annotation, *restofheader, *ptr, *nextptr;
bool allocp;
Chrpos_T chrpos, intervalend;
annotation = IIT_annotation(&restofheader,iit,indexi,&allocp);
interval = IIT_interval(iit,indexi);
chrpos = Interval_low(interval);
intervalend = Interval_high(interval);
ptr = annotation;
if (zeroesp == true) {
while (*lastcoord < chrpos) {
if (statsp == false) {
printf("%s\t%u\t%d\t\n",divstring,*lastcoord,0);
}
(*lastcoord)++;
}
}
while (chrpos < coordstart) {
if ((ptr = index(ptr,'\n')) == NULL) {
fprintf(stderr,"Premature end of tally from %u to %u\n",
Interval_low(interval),Interval_high(interval));
return total;
} else {
ptr++;
}
chrpos++;
}
while (chrpos <= intervalend && chrpos <= coordend) {
subtotal = 0;
nextptr = get_total_tally(&subtotal,ptr);
if (subtotal > 0 || zeroesp == true) {
if (statsp == false) {
printf("%s\t%u\t%ld\t",divstring,chrpos,total);
print_line(ptr);
printf("\n");
}
}
total += subtotal;
ptr = nextptr;
if ((ptr = index(ptr,'\n')) == NULL) {
fprintf(stderr,"Premature end of tally from %u to %u\n",
Interval_low(interval),Interval_high(interval));
return total;
} else {
ptr++;
}
chrpos++;
}
*lastcoord = chrpos;
if (allocp == true) {
FREE(restofheader);
}
return total;
}
/* Need to store just the part of the query specified (e.g., 1..10) */
static void
print_interval_centered (char *divstring, Chrpos_T coordstart, int index, IIT_T iit, int fieldint) {
Interval_T interval;
char *label, *annotation, *restofheader, centerchar;
bool allocp;
int annotlength, left, centerpos;
if (fieldint < 0) {
annotation = IIT_annotation(&restofheader,iit,index,&allocp);
if (allocp == true) {
FREE(restofheader);
}
} else {
annotation = IIT_fieldvalue(iit,index,fieldint);
allocp = true;
}
annotlength = strlen(annotation);
if (annotation[annotlength-1] == '\n') {
annotlength--;
}
interval = IIT_interval(iit,index);
left = coordstart - Interval_low(interval); /* + length(query) - queryend */
if (Interval_sign(interval) < 0) {
centerpos = annotlength-left-1;
} else {
centerpos = left;
}
centerchar = annotation[centerpos];
if (centeruc == true && islower(centerchar)) {
if (fieldint >= 0 && allocp == true) {
FREE(annotation);
}
} else {
print_spaces(centerlength-left);
if (Interval_sign(interval) < 0) {
print_complement(annotation,annotlength-1,centerpos+1);
printf("[%c]",complCode[(int) centerchar]);
print_complement(annotation,centerpos-1,0);
} else {
print_forward(annotation,0,centerpos-1);
printf("[%c]",centerchar);
print_forward(annotation,centerpos+1,annotlength-1);
}
print_spaces(centerlength+left-annotlength);
if (fieldint >= 0 && allocp == true) {
FREE(annotation);
}
printf("\t");
if (Interval_type(interval) > 0) {
printf("%s\t",IIT_typestring(iit,Interval_type(interval)));
}
if (divstring != NULL) {
if (Interval_sign(interval) < 0) {
printf("-%s:",divstring);
} else {
printf("+%s:",divstring);
}
}
if (signedp == false) {
printf("%u..%u",Interval_low(interval),Interval_high(interval));
} else if (Interval_sign(interval) < 0) {
printf("%u..%u",Interval_high(interval),Interval_low(interval));
} else {
printf("%u..%u",Interval_low(interval),Interval_high(interval));
}
printf("\t");
label = IIT_label(iit,index,&allocp);
printf("%s",label);
if (allocp == true) {
FREE(label);
}
printf("\n");
}
return;
}
bool
Parserange_universal_iit (char **div, bool *revcomp,
Genomicpos_T *genomicstart, Genomicpos_T *genomiclength,
Genomicpos_T *chrstart, Genomicpos_T *chrend,
Genomicpos_T *chroffset, Genomicpos_T *chrlength,
char *query, IIT_T chromosome_iit, IIT_T contig_iit) {
char *coords;
Genomicpos_T result, left, length;
Interval_T interval;
int theindex;
int rc;
*revcomp = false;
if (index(query,':')) {
/* Segment must be a genome, chromosome, or contig */
debug(printf("Parsed query %s into ",query));
*div = strtok(query,":");
if ((*div)[0] == '+') {
*revcomp = false;
*div = &((*div)[1]);
} else if ((*div)[0] == '_') {
*revcomp = true;
*div = &((*div)[1]);
}
coords = strtok(NULL,":");
debug(printf("segment %s and coords %s\n",*div,coords));
debug(printf("Interpreting segment %s as a chromosome\n",*div));
if (coords == NULL) {
debug(printf(" entire chromosome\n"));
rc = translate_chromosomepos_universal(&(*genomicstart),&(*genomiclength),*div,left=0,length=0,chromosome_iit);
} else if (isnumberp(&result,coords)) {
debug(printf(" and coords %s as a number\n",coords));
rc = translate_chromosomepos_universal(&(*genomicstart),&(*genomiclength),*div,left=result-1,length=1,chromosome_iit);
} else if (isrange(&left,&length,&(*revcomp),coords)) {
debug(printf(" and coords %s as a range starting at %u with length %u and revcomp = %d\n",
coords,left,length,*revcomp));
rc = translate_chromosomepos_universal(&(*genomicstart),&(*genomiclength),*div,left,length,chromosome_iit);
} else {
debug(printf(" but coords %s is neither a number nor a range\n",coords));
rc = -1;
}
/* Compute chromosomal coordinates */
*chrstart = left;
*chrend = *chrstart + *genomiclength;
*chrstart += 1U; /* Make 1-based */
/* Get chromosomal information */
if ((theindex = IIT_find_one(chromosome_iit,*div)) < 0) {
fprintf(stderr,"Cannot find chromosome %s in chromosome IIT file\n",*div);
/* exit(9); */
} else {
interval = IIT_interval(chromosome_iit,theindex);
*chroffset = Interval_low(interval);
*chrlength = Interval_length(interval);
}
if (rc != 0) {
/* Try contig */
debug(printf("Interpreting segment %s as a contig\n",*div));
if (isnumberp(&result,coords)) {
debug(printf(" and coords %s as a number\n",coords));
rc = translate_contig(&(*genomicstart),&(*genomiclength),*div,left=result-1,length=1,contig_iit);
} else if (isrange(&left,&length,&(*revcomp),coords)) {
debug(printf(" and coords %s as a range starting at %u with length %u and revcomp = %d\n",
coords,left,length,*revcomp));
rc = translate_contig(&(*genomicstart),&(*genomiclength),*div,left,length,contig_iit);
} else {
debug(printf(" but coords %s is neither a number nor a range\n",coords));
rc = -1;
}
}
if (rc != 0) {
fprintf(stderr,"Can't find coordinates %s:%s\n",*div,coords);
return false;
} else {
return true;
}
} else {
/* Query must be a genomic position, genomic range, or contig */
*chrstart = *chroffset = *chrlength = 0;
debug(printf("Parsed query %s as atomic ",query));
if (isnumberp(&result,query)) {
debug(printf("number\n"));
*genomicstart = result-1;
*genomiclength = 1;
} else if (isrange(&left,&length,&(*revcomp),query)) {
debug(printf("range\n"));
*genomicstart = left;
*genomiclength = length;
} else {
debug(printf("contig\n"));
return false;
#if 0
rc = translate_contig(&(*genomicstart),&(*genomiclength),query,left=0,length=0,contig_iit);
IIT_free(&contig_iit);
#endif
}
*div = convert_to_chrpos_iit(&(*chrstart),chromosome_iit,*genomicstart);
*chrend = *chrstart + *genomiclength;
*chrstart += 1U; /* Make 1-based */
/* Try chromosome first */
if ((theindex = IIT_find_one(chromosome_iit,*div)) < 0) {
fprintf(stderr,"Cannot find chromosome %s in chromosome IIT file\n",*div);
return false;
} else {
interval = IIT_interval(chromosome_iit,theindex);
*chroffset = Interval_low(interval);
*chrlength = Interval_length(interval);
return true;
}
}
}
bool
Parserange_universal (char **div, bool *revcomp,
Genomicpos_T *genomicstart, Genomicpos_T *genomiclength,
Genomicpos_T *chrstart, Genomicpos_T *chrend,
Genomicpos_T *chroffset, Genomicpos_T *chrlength,
char *query, char *genomesubdir, char *fileroot) {
char *coords, *filename;
Genomicpos_T result, left, length;
IIT_T chromosome_iit, contig_iit;
Interval_T interval;
int theindex;
int rc;
*revcomp = false;
if (index(query,':')) {
/* Segment must be a genome, chromosome, or contig */
debug(printf("Parsed query %s into ",query));
*div = strtok(query,":");
if ((*div)[0] == '+') {
*revcomp = false;
*div = &((*div)[1]);
} else if ((*div)[0] == '_') {
*revcomp = true;
*div = &((*div)[1]);
}
coords = strtok(NULL,":");
debug(printf("segment %s and coords %s\n",*div,coords));
/* Try chromosome first */
filename = (char *) CALLOC(strlen(genomesubdir)+strlen("/")+strlen(fileroot)+
strlen(".chromosome.iit")+1,sizeof(char));
sprintf(filename,"%s/%s.chromosome.iit",genomesubdir,fileroot);
chromosome_iit = IIT_read(filename,/*name*/NULL,/*readonlyp*/true,/*divread*/READ_ALL,
/*divstring*/NULL,/*add_iit_p*/false,/*labels_read_p*/true);
FREE(filename);
debug(printf("Interpreting segment %s as a chromosome\n",*div));
if (coords == NULL) {
debug(printf(" entire chromosome\n"));
rc = translate_chromosomepos_universal(&(*genomicstart),&(*genomiclength),*div,left=0,length=0,chromosome_iit);
} else if (isnumberp(&result,coords)) {
debug(printf(" and coords %s as a number\n",coords));
rc = translate_chromosomepos_universal(&(*genomicstart),&(*genomiclength),*div,left=result-1,length=1,chromosome_iit);
} else if (isrange(&left,&length,&(*revcomp),coords)) {
debug(printf(" and coords %s as a range starting at %u with length %u and revcomp = %d\n",
coords,left,length,*revcomp));
rc = translate_chromosomepos_universal(&(*genomicstart),&(*genomiclength),*div,left,length,chromosome_iit);
} else {
debug(printf(" but coords %s is neither a number nor a range\n",coords));
rc = -1;
}
/* Compute chromosomal coordinates */
*chrstart = left;
*chrend = *chrstart + *genomiclength;
*chrstart += 1U; /* Make 1-based */
/* Get chromosomal information */
if ((theindex = IIT_find_one(chromosome_iit,*div)) < 0) {
fprintf(stderr,"Cannot find chromosome %s in chromosome IIT file\n",*div);
/* exit(9); */
} else {
interval = IIT_interval(chromosome_iit,theindex);
*chroffset = Interval_low(interval);
*chrlength = Interval_length(interval);
}
IIT_free(&chromosome_iit);
#if 0
/* Contig IIT's are of type 1, which require some work to compute
on current div-based scheme. Just abandoning for now. */
if (rc != 0) {
/* Try contig */
filename = (char *) CALLOC(strlen(genomesubdir)+strlen("/")+strlen(fileroot)+
strlen(".contig.iit")+1,sizeof(char));
sprintf(filename,"%s/%s.contig.iit",genomesubdir,fileroot);
contig_iit = IIT_read(filename,/*name*/NULL,/*readonlyp*/true,/*divread*/READ_ALL,
/*divstring*/NULL,/*add_iit_p*/false,/*labels_read_p*/true);
FREE(filename);
debug(printf("Interpreting segment %s as a contig\n",*div));
if (coords == NULL) {
debug(printf(" entire contig\n"));
rc = translate_contig_universal(&(*genomicstart),&(*genomiclength),*div,left=0,length=0,chromosome_iit);
} else if (isnumberp(&result,coords)) {
debug(printf(" and coords %s as a number\n",coords));
rc = translate_contig(&(*genomicstart),&(*genomiclength),*div,left=result-1,length=1,contig_iit);
} else if (isrange(&left,&length,&(*revcomp),coords)) {
debug(printf(" and coords %s as a range starting at %u with length %u and revcomp = %d\n",
coords,left,length,*revcomp));
rc = translate_contig(&(*genomicstart),&(*genomiclength),*div,left,length,contig_iit);
} else {
debug(printf(" but coords %s is neither a number nor a range\n",coords));
rc = -1;
}
IIT_free(&contig_iit);
}
#endif
if (rc != 0) {
fprintf(stderr,"Can't find coordinates %s:%s\n",*div,coords);
return false;
} else {
return true;
}
} else {
/* Query must be a genomic position, genomic range, or contig */
*chrstart = *chroffset = *chrlength = 0;
debug(printf("Parsed query %s as atomic ",query));
if (isnumberp(&result,query)) {
debug(printf("number\n"));
*genomicstart = result-1;
*genomiclength = 1;
} else if (isrange(&left,&length,&(*revcomp),query)) {
debug(printf("range\n"));
*genomicstart = left;
*genomiclength = length;
} else {
debug(printf("contig\n"));
return false;
#if 0
filename = (char *) CALLOC(strlen(genomesubdir)+strlen("/")+strlen(fileroot)+
strlen(".contig.iit")+1,sizeof(char));
sprintf(filename,"%s/%s.contig.iit",genomesubdir,fileroot);
contig_iit = IIT_read(filename,/*name*/NULL,/*readonlyp*/true,/*divread*/READ_ALL,
/*divstring*/NULL,/*add_iit_p*/false,/*labels_read_p*/true);
FREE(filename);
rc = translate_contig(&(*genomicstart),&(*genomiclength),query,left=0,length=0,contig_iit);
IIT_free(&contig_iit);
#endif
}
*div = convert_to_chrpos(&(*chrstart),genomesubdir,fileroot,*genomicstart);
*chrend = *chrstart + *genomiclength;
*chrstart += 1U; /* Make 1-based */
/* Try chromosome first */
filename = (char *) CALLOC(strlen(genomesubdir)+strlen("/")+strlen(fileroot)+
strlen(".chromosome.iit")+1,sizeof(char));
sprintf(filename,"%s/%s.chromosome.iit",genomesubdir,fileroot);
chromosome_iit = IIT_read(filename,/*name*/NULL,/*readonlyp*/true,/*divread*/READ_ALL,
/*divstring*/NULL,/*add_iit_p*/false,/*labels_read_p*/true);
FREE(filename);
if ((theindex = IIT_find_one(chromosome_iit,*div)) < 0) {
fprintf(stderr,"Cannot find chromosome %s in chromosome IIT file\n",*div);
IIT_free(&chromosome_iit);
return false;
} else {
interval = IIT_interval(chromosome_iit,theindex);
*chroffset = Interval_low(interval);
*chrlength = Interval_length(interval);
IIT_free(&chromosome_iit);
return true;
}
}
}