34dc2bb14c176fb8f788df267fdb3fd4c8f461b4
braney
Fri Feb 17 07:00:46 2023 -0800
fix some encoding issues and an errant printf on the TOGO hgc page.
diff --git src/hg/hgc/togaClick.c src/hg/hgc/togaClick.c
index 5608e5c..00a4c8c 100644
--- src/hg/hgc/togaClick.c
+++ src/hg/hgc/togaClick.c
@@ -1,696 +1,695 @@
/* togaClick - click handling for TOGA tracks */
#include "common.h"
#include "hgc.h"
#include "togaClick.h"
#include "string.h"
#include "htmshell.h"
#include "chromAlias.h"
struct togaDataBB *togaDataBBLoad(char **row)
/* Load a togaData from row fetched with select * from togaData
* from database. Dispose of this with togaDataFree(). */
{
struct togaDataBB *ret;
AllocVar(ret);
ret->projection = cloneString(row[0]);
ret->ref_trans_id = cloneString(row[1]);
ret->ref_region = cloneString(row[2]);
ret->query_region = cloneString(row[3]);
ret->chain_score = cloneString(row[4]);
ret->chain_synteny = cloneString(row[5]);
ret->chain_flank = cloneString(row[6]);
ret->chain_gl_cds_fract = cloneString(row[7]);
ret->chain_loc_cds_fract = cloneString(row[8]);
ret->chain_exon_cov = cloneString(row[9]);
ret->chain_intron_cov = cloneString(row[10]);
ret->status = cloneString(row[11]);
ret->perc_intact_ign_M = cloneString(row[12]);
ret->perc_intact_int_M = cloneString(row[13]);
ret->intact_codon_prop = cloneString(row[14]);
ret->ouf_prop = cloneString(row[15]);
ret->mid_intact = cloneString(row[16]);
ret->mid_pres = cloneString(row[17]);
ret->prot_alignment = cloneString(row[18]);
ret->svg_line = cloneString(row[19]);
ret->ref_link = cloneString(row[20]);
ret->inact_mut_html_table = cloneString(row[21]);
ret->exon_ali_html = cloneString(row[22]);
return ret;
}
struct togaData *togaDataLoad(char **row)
/* Load a togaData from row fetched with select * from togaData
* from database. Dispose of this with togaDataFree(). */
{
struct togaData *ret;
AllocVar(ret);
ret->projection = cloneString(row[0]);
ret->ref_trans_id = cloneString(row[1]);
ret->ref_region = cloneString(row[2]);
ret->query_region = cloneString(row[3]);
ret->chain_score = cloneString(row[4]);
ret->chain_synteny = cloneString(row[5]);
ret->chain_flank = cloneString(row[6]);
ret->chain_gl_cds_fract = cloneString(row[7]);
ret->chain_loc_cds_fract = cloneString(row[8]);
ret->chain_exon_cov = cloneString(row[9]);
ret->chain_intron_cov = cloneString(row[10]);
ret->status = cloneString(row[11]);
ret->perc_intact_ign_M = cloneString(row[12]);
ret->perc_intact_int_M = cloneString(row[13]);
ret->intact_codon_prop = cloneString(row[14]);
ret->ouf_prop = cloneString(row[15]);
ret->mid_intact = cloneString(row[16]);
ret->mid_pres = cloneString(row[17]);
ret->prot_alignment = cloneString(row[18]);
ret->svg_line = cloneString(row[19]);
return ret;
}
void togaDataBBFree(struct togaDataBB **pEl)
/* Free a single dynamically allocated togaDatasuch as created
* with togaDataLoad(). */
{
struct togaDataBB *el;
if ((el = *pEl) == NULL) return;
freeMem(el->projection);
freeMem(el->ref_trans_id);
freeMem(el->ref_region);
freeMem(el->query_region);
freeMem(el->chain_score);
freeMem(el->chain_synteny);
freeMem(el->chain_flank);
freeMem(el->chain_gl_cds_fract);
freeMem(el->chain_loc_cds_fract);
freeMem(el->chain_exon_cov);
freeMem(el->chain_intron_cov);
freeMem(el->status);
freeMem(el->perc_intact_ign_M);
freeMem(el->perc_intact_int_M);
freeMem(el->intact_codon_prop);
freeMem(el->ouf_prop);
freeMem(el->mid_intact);
freeMem(el->mid_pres);
freeMem(el->prot_alignment);
freeMem(el->svg_line);
freeMem(el->ref_link);
freeMem(el->inact_mut_html_table);
freeMem(el->exon_ali_html);
freez(pEl);
}
void togaDataFree(struct togaData **pEl)
/* Free a single dynamically allocated togaDatasuch as created
* with togaDataLoad(). */
{
struct togaData *el;
if ((el = *pEl) == NULL) return;
freeMem(el->projection);
freeMem(el->ref_trans_id);
freeMem(el->ref_region);
freeMem(el->query_region);
freeMem(el->chain_score);
freeMem(el->chain_synteny);
freeMem(el->chain_flank);
freeMem(el->chain_gl_cds_fract);
freeMem(el->chain_loc_cds_fract);
freeMem(el->chain_exon_cov);
freeMem(el->chain_intron_cov);
freeMem(el->status);
freeMem(el->perc_intact_ign_M);
freeMem(el->perc_intact_int_M);
freeMem(el->intact_codon_prop);
freeMem(el->ouf_prop);
freeMem(el->mid_intact);
freeMem(el->mid_pres);
freeMem(el->prot_alignment);
freeMem(el->svg_line);
freez(pEl);
}
struct togaNucl *togaNuclLoad(char **row)
/* Load a togaNucl from row fetched with select * from togaNucl
* from database. Dispose of this with togaNuclFree(). */
{
struct togaNucl *ret;
AllocVar(ret);
ret->transcript = cloneString(row[0]);
ret->exon_num = cloneString(row[1]);
ret->exon_region = cloneString(row[2]);
ret->pid = cloneString(row[3]);
ret->blosum = cloneString(row[4]);
ret->gaps = cloneString(row[5]);
ret->ali_class = cloneString(row[6]);
ret->exp_region = cloneString(row[7]);
ret->in_exp_region = cloneString(row[8]);
ret->alignment = cloneString(row[9]);
return ret;
}
void togaNuclFree(struct togaNucl **pEl)
/* Free a single dynamically allocated togaNucl such as created
* with togaNuclLoad(). */
{
struct togaNucl *el;
if ((el = *pEl) == NULL) return;
freeMem(el->transcript);
freeMem(el->exon_num);
freeMem(el->exon_region);
freeMem(el->pid);
freeMem(el->blosum);
freeMem(el->gaps);
freeMem(el->ali_class);
freeMem(el->exp_region);
freeMem(el->in_exp_region);
freeMem(el->alignment);
freez(pEl);
}
struct togaInactMut *togaInactMutLoad(char **row)
/* Load a togaInactMut from row fetched with select * from togaInactMut
* from database. Dispose of this with togaInactMutFree(). */
{
struct togaInactMut *ret;
AllocVar(ret);
ret->transcript = cloneString(row[0]);
ret->exon_num = cloneString(row[1]);
ret->position = cloneString(row[2]);
ret->mut_class = cloneString(row[3]);
ret->mutation = cloneString(row[4]);
ret->is_inact = cloneString(row[5]);
ret->mut_id = cloneString(row[6]);
return ret;
}
void togaInactMutFree(struct togaInactMut **pEl)
/* Free a single dynamically allocated togaInactMut such as created
* with togaInactMutLoad(). */
{
struct togaInactMut *el;
if ((el = *pEl) == NULL) return;
freeMem(el->transcript);
freeMem(el->exon_num);
freeMem(el->position);
freeMem(el->mut_class);
freeMem(el->mutation);
freeMem(el->is_inact);
freeMem(el->mut_id);
freez(pEl);
}
void extractHLTOGAsuffix(char *suffix)
/* Extract suffix from TOGA table name.
Prefix must be HLTOGAannot */
{
int suff_len = strlen(suffix);
if (suff_len <= HLTOGA_BED_PREFIX_LEN)
// we cannot chop first PREFIX_LEN characters
{
// TODO: NOT SURE IF IT WORKS; but this must not happen
char empty[5] = { '\0' };
strcpy(suffix, empty);
} else {
// just start the string 11 characters upstream
memmove(suffix, suffix + HLTOGA_BED_PREFIX_LEN, suff_len - HLTOGA_BED_PREFIX_LEN + 1);
}
}
void HLprintQueryProtSeqForAli(char *proteinAlignment) {
// take protein sequence alignment
// print only the query sequence
char *str = proteinAlignment;
int printed_char_num = 0;
while ((str = strstr(str, "que:")) != NULL)
{
str += 10;
char ch;
while ((ch = *str++) != '<') {
if (ch != '-') {
putchar(ch);
++printed_char_num;
}
if (printed_char_num == 80) {
printed_char_num = 0;
printf(" ");
}
}
}
}
void doHillerLabTOGAGeneBig(char *database, struct trackDb *tdb, char *item, char *table_name)
/* Put up TOGA Gene track info. */
// To think about -> put into a single bigBed
// string: HTML formatted inact mut
// string: HTML formatted exon ali section
{
int start = cartInt(cart, "o");
int end = cartInt(cart, "t");
char *chrom = cartString(cart, "c");
char *fileName = bbiNameFromSettingOrTable(tdb, NULL, tdb->table);
struct bbiFile *bbi = bigBedFileOpenAlias(hReplaceGbdb(fileName), chromAliasFindAliases);
struct lm *lm = lmInit(0);
struct bigBedInterval *bbList = bigBedIntervalQuery(bbi, chrom, start, end, 0, lm);
struct bigBedInterval *bb;
char *fields[bbi->fieldCount];
for (bb = bbList; bb != NULL; bb = bb->next)
{
if (!(bb->start == start && bb->end == end))
continue;
// our names are unique
char *name = cloneFirstWordByDelimiterNoSkip(bb->rest, '\t');
boolean match = (isEmpty(name) && isEmpty(item)) || sameOk(name, item);
if (!match)
continue;
char startBuf[16], endBuf[16];
bigBedIntervalToRow(bb, chrom, startBuf, endBuf, fields, bbi->fieldCount);
break;
}
printf("
Projection %s
\n", item);
struct togaDataBB *info = togaDataBBLoad(&fields[11]); // Bogdan: why 11? 0-11 are bed-like fields likely
printf("Reference transcript: %s ", info->ref_link);
printf("Genomic locus in reference: %s \n", info->ref_region);
printf("Genomic locus in query: %s \n", info->query_region);
printf("Projection classification: %s \n", info->status);
printf("Probability that query locus is orthologous: %s \n", info->chain_score);
// list of chain features (for orthology classification)
printf("Show features used for ortholog probability\n");
printf("
\n");
printf("
\n");
printf("
Synteny (log10 value): %s
\n", info->chain_synteny);
printf("
Global CDS fraction: %s
\n", info->chain_gl_cds_fract);
printf("
Local CDS fraction: %s
\n", info->chain_loc_cds_fract);
printf("
Local intron fraction: %s
\n", info->chain_intron_cov);
printf("
Local CDS coverage: %s
\n", info->chain_exon_cov);
printf("
Flank fraction: %s
\n", info->chain_flank);
printf("
\n");
printf(" \nFeature description:\n");
printf("For each projection (one reference transcript and one overlapping chain),\n");
printf("TOGA computes the following features by intersecting the reference coordinates of aligning\n");
printf("blocks in the chain with different gene parts (coding exons, UTR (untranslated region) exons, introns)\n");
printf("and the respective intergenic regions.\n \n");
printf("We define the following variables:\n
\n");
printf("
c: number of reference bases in the intersection between chain blocks and coding exons of the gene under consideration.
\n");
printf("
C: number of reference bases in the intersection between chain blocks and coding exons of all genes.
\n");
printf("
a: number of reference bases in the intersection between chain blocks and coding exons and introns of the gene under consideration.
\n");
printf("
A: number of reference bases in the intersection between chain blocks and coding exons and introns of all genes and the intersection\n");
printf("between chain blocks and intergenic regions (excludes UTRs).
\n");
printf("
f: number of reference bases in chain blocks overlapping the 10 kb flanks of the gene under consideration.\n");
printf("Alignment blocks overlapping exons of another gene that is located in these 10 kb flanks are ignored.
\n");
printf("
i: number of reference bases in the intersection between chain blocks and introns of the gene under consideration.
\n");
printf("
CDS (coding sequence): length of the coding region of the gene under consideration.
\n");
printf("
I: sum of all intron lengths of the gene under consideration.
\n");
printf("
\n");
printf("Using these variables, TOGA computes the following features:\n");
printf("
\n");
-printf("
“global CDS fraction” as C / A. Chains with a high value have alignments that largely overlap coding exons,");
+printf("
"global CDS fraction" as C / A. Chains with a high value have alignments that largely overlap coding exons,");
printf("which is a hallmark of paralogous or processed pseudogene chains. In contrast, chains with a low value also align many ");
printf("intronic and intergenic regions, which is a hallmark of orthologous chains.
\n");
-printf("
“local CDS fraction” as c / a. Orthologous chains tend to have a lower value, as intronic ");
+printf("
"local CDS fraction" as c / a. Orthologous chains tend to have a lower value, as intronic ");
printf("regions partially align. This feature is not computed for single-exon genes.
\n");
-printf("
“local intron fraction” as i / I. Orthologous chains tend to have a higher value.");
+printf("
"local intron fraction" as i / I. Orthologous chains tend to have a higher value.");
printf("This feature is not computed for single-exon genes.
\n");
-printf("
“flank fraction” as f / 20,000. Orthologous chains tend to have higher values,");
+printf("
"flank fraction" as f / 20,000. Orthologous chains tend to have higher values,");
printf("as flanking intergenic regions partially align. This feature is important to detect orthologous loci of single-exon genes.
\n");
-printf("
“synteny” as log10 of the number of genes, whose coding exons overlap by at least one base aligning");
+printf("
"synteny" as log10 of the number of genes, whose coding exons overlap by at least one base aligning");
printf("blocks of this chain. Orthologous chains tend to cover several genes located in a conserved order, resulting in higher synteny values.
\n");
-printf("
“local CDS coverage” as c / CDS, which is only used for single-exon genes.
\n");
+printf("
"local CDS coverage" as c / CDS, which is only used for single-exon genes.
\n");
printf("
\n");
printf("\n
\n \n");
htmlHorizontalLine();
// show inact mut plot
printf("
Visualization of inactivating mutations on exon-intron structure
\n");
printf("%s \n", info->svg_line);
printf(" Exons shown in grey are missing (often overlap assembly gaps).\nExons shown in");
printf(" red or blue are deleted or do not align at all.\nRed indicates that the exon deletion ");
printf("shifts the reading frame, while blue indicates that exon deletion(s) are framepreserving. \n");
// GLP features
printf("Show features used for transcript classification\n");
printf("
\n");
printf("
\n");
printf("
Percent intact, ignoring missing sequence: %s
\n", info->perc_intact_ign_M);
printf("
Percent intact, treating missing as intact sequence: %s
\n", info->perc_intact_int_M);
printf("
Proportion of intact codons: %s
\n", info->intact_codon_prop);
printf("
Percent of CDS not covered by this chain (0 unless the chain covers only a part of the gene): %s
\n", info->ouf_prop);
if (sameWord(info->mid_intact, ONE_))
{
printf("
Middle 80 percent of CDS intact: %s
\n", YES_);
} else {
printf("
Middle 80 percent of CDS intact: %s
\n", NO_);
}
if (sameWord(info->mid_pres, ONE_))
{
printf("
\n");
// printf("{protein seq of the query without dashes or other things. Should end with *}\n");
printf("");
HLprintQueryProtSeqForAli(info->prot_alignment);
printf("\n \n\n
\n");
// and show protein sequence
htmlHorizontalLine();
printf("
\n", item);
char query[256];
struct sqlResult *sr = NULL;
char **row;
struct togaData *info = NULL;
sqlSafef(query, sizeof(query), "select * from %s where transcript='%s'", togaDataTableName, item);
sr = sqlGetResult(conn, query);
if ((row = sqlNextRow(sr)) != NULL) {
info = togaDataLoad(row); // parse sql output
// fill HTML template:
printf("Reference transcript: %s ",
info->ref_trans_id, info->ref_trans_id);
printf("Genomic locus in reference: %s \n", info->ref_region);
printf("Genomic locus in query: %s \n", info->query_region);
printf("Projection classification: %s \n", info->status);
printf("Probability that query locus is orthologous: %s \n", info->chain_score);
// list of chain features (for orthology classification)
printf("Show features used for ortholog probability\n");
printf("
\n");
printf("
\n");
printf("
Synteny (log10 value): %s
\n", info->chain_synteny);
printf("
Global CDS fraction: %s
\n", info->chain_gl_cds_fract);
printf("
Local CDS fraction: %s
\n", info->chain_loc_cds_fract);
printf("
Local intron fraction: %s
\n", info->chain_intron_cov);
printf("
Local CDS coverage: %s
\n", info->chain_exon_cov);
printf("
Flank fraction: %s
\n", info->chain_flank);
printf("
\n");
printf(" \nFeature description:\n");
printf("For each projection (one reference transcript and one overlapping chain),\n");
printf("TOGA computes the following features by intersecting the reference coordinates of aligning\n");
printf("blocks in the chain with different gene parts (coding exons, UTR (untranslated region) exons, introns)\n");
printf("and the respective intergenic regions.\n \n");
printf("We define the following variables:\n
\n");
printf("
c: number of reference bases in the intersection between chain blocks and coding exons of the gene under consideration.
\n");
printf("
C: number of reference bases in the intersection between chain blocks and coding exons of all genes.
\n");
printf("
a: number of reference bases in the intersection between chain blocks and coding exons and introns of the gene under consideration.
\n");
printf("
A: number of reference bases in the intersection between chain blocks and coding exons and introns of all genes and the intersection\n");
printf("between chain blocks and intergenic regions (excludes UTRs).
\n");
printf("
f: number of reference bases in chain blocks overlapping the 10 kb flanks of the gene under consideration.\n");
printf("Alignment blocks overlapping exons of another gene that is located in these 10 kb flanks are ignored.
\n");
printf("
i: number of reference bases in the intersection between chain blocks and introns of the gene under consideration.
\n");
printf("
CDS (coding sequence): length of the coding region of the gene under consideration.
\n");
printf("
I: sum of all intron lengths of the gene under consideration.
\n");
printf("
\n");
printf("Using these variables, TOGA computes the following features:\n");
printf("
\n");
- printf("
“global CDS fraction” as C / A. Chains with a high value have alignments that largely overlap coding exons,");
+ printf("
"global CDS fraction" as C / A. Chains with a high value have alignments that largely overlap coding exons,");
printf("which is a hallmark of paralogous or processed pseudogene chains. In contrast, chains with a low value also align many ");
printf("intronic and intergenic regions, which is a hallmark of orthologous chains.
\n");
- printf("
“local CDS fraction” as c / a. Orthologous chains tend to have a lower value, as intronic ");
+ printf("
"local CDS fraction" as c / a. Orthologous chains tend to have a lower value, as intronic ");
printf("regions partially align. This feature is not computed for single-exon genes.
\n");
- printf("
“local intron fraction” as i / I. Orthologous chains tend to have a higher value.");
+ printf("
"local intron fraction" as i / I. Orthologous chains tend to have a higher value.");
printf("This feature is not computed for single-exon genes.
\n");
- printf("
“flank fraction” as f / 20,000. Orthologous chains tend to have higher values,");
+ printf("
"flank fraction" as f / 20,000. Orthologous chains tend to have higher values,");
printf("as flanking intergenic regions partially align. This feature is important to detect orthologous loci of single-exon genes.
\n");
- printf("
“synteny” as log10 of the number of genes, whose coding exons overlap by at least one base aligning");
+ printf("
"synteny" as log10 of the number of genes, whose coding exons overlap by at least one base aligning");
printf("blocks of this chain. Orthologous chains tend to cover several genes located in a conserved order, resulting in higher synteny values.
\n");
- printf("
“local CDS coverage” as c / CDS, which is only used for single-exon genes.
\n");
+ printf("
"local CDS coverage" as c / CDS, which is only used for single-exon genes.
\n");
printf("
\n");
printf("\n
\n \n");
htmlHorizontalLine();
// show inact mut plot
printf("
Visualization of inactivating mutations on exon-intron structure
\n");
printf("%s \n", info->svg_line);
printf(" Exons shown in grey are missing (often overlap assembly gaps).\nExons shown in");
printf(" red or blue are deleted or do not align at all.\nRed indicates that the exon deletion ");
printf("shifts the reading frame, while blue indicates that exon deletion(s) are framepreserving. \n");
// GLP features
printf("Show features used for transcript classification\n");
printf("
\n");
printf("
\n");
printf("
Percent intact, ignoring missing sequence: %s
\n", info->perc_intact_ign_M);
printf("
Percent intact, treating missing as intact sequence: %s
\n", info->perc_intact_int_M);
printf("
Proportion of intact codons: %s
\n", info->intact_codon_prop);
printf("
Percent of CDS not covered by this chain (0 unless the chain covers only a part of the gene): %s
\n", info->ouf_prop);
if (sameWord(info->mid_intact, ONE_))
{
printf("
Middle 80 percent of CDS intact: %s
\n", YES_);
} else {
printf("
Middle 80 percent of CDS intact: %s
\n", NO_);
}
if (sameWord(info->mid_pres, ONE_))
{
printf("
\n");
// do not forget to free toga data struct
togaDataFree(&info);
} else {
// no data found, need to report this
printf("Not found data for %s\n", item);
}
sqlFreeResult(&sr);
}
// show inactivating mutations if required
printf("
List of inactivating mutations
\n");
if (hTableExists(database, togaInactMutTableName))
{
char query[256];
struct sqlResult *sr = NULL;
char **row;
sqlSafef(query, sizeof(query), "select * from %s where transcript='%s'", togaInactMutTableName, item);
sr = sqlGetResult(conn, query);
printf("Show inactivating mutations\n");
printf("
\n");
printf("
\n"); // init table
printf("
Exon number
Codon number
Mutation class
Mutation
Treated as inactivating
Mutation ID
\n");
printf("
\n");
while ((row = sqlNextRow(sr)) != NULL)
{
struct togaInactMut *info = NULL;
info = togaInactMutLoad(row);
printf("
\n");
printf("
%s
\n", info->exon_num);
printf("
%s
\n", info->position);
printf("
%s
\n", info->mut_class);
printf("
%s
\n", info->mutation);
if (sameWord(info->is_inact, ONE_)){
printf("
\n");
} else {
printf("Sorry, cannot find TOGANucl table. \n");
}
htmlHorizontalLine();
// TODO: check whether I need this
printf("%s", hgTracksPathAndSettings());
hPrintf("");
hPrintf("");
hPrintf("");
printTrackHtml(tdb); // and do I need this?
hFreeConn(&conn);
}