\n", target);
+}
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 v2 %s
\n", item);
struct togaDataBB *info = togaDataBBLoad(&fields[11], bbi->fieldCount); // 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");
+panelPrompt("collapseChain", "Show features used for ortholog probability");
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("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("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("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("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("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("
\n");
printf("
\n");
printf("
\n");
// 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");
+panelPrompt("collapseGLP", "Show features used for transcript classification");
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("- Middle 80 percent of CDS present: %s
\n", YES_);
} else {
printf("- Middle 80 percent of CDS present: %s
\n", NO_);
}
if (info->numExonsMutated != NULL && info->percentExonsMutated != NULL) {
printf("- Number of exons with inactivating mutations: %s (%s%% of the present exons; threshold is 20%%)
\n", info->numExonsMutated, info->percentExonsMutated);
}
printf("
\n
\n");
printf("
\n");
printf("
Predicted protein sequence
\n");
-printf("
Show protein sequence of query\n");
-printf("
\n");
+panelPrompt("collapseProt", "Show protein sequence of query");
printf("
");
HLprintQueryProtSeqForAli(info->prot_alignment);
printf("\n
\n
\n");
if (info->protseqFrameCorrected != NULL) {
- printf("
Show frame-corrected protein sequence of query (potential frameshifts are masked)\n");
- printf("
\n");
+ panelPrompt("collapseProtFrameCorrected", "Show frame-corrected protein sequence of query (potential frameshifts are masked)");
printf("
");
print_with_newlines(info->protseqFrameCorrected);
printf("\n
\n
\n");
}
if (info->CDSseq != NULL) {
printf("
\n");
printf("
Predicted coding (DNA) sequence
\n");
- printf("
Show coding sequence of query\n");
- printf("
\n");
+ panelPrompt("collapseCDS", "Show coding sequence of query");
printf("
");
print_with_newlines(info->CDSseq);
printf("\n
\n
\n");
}
// and show protein sequence
printf("
\n");
printf("
Protein sequence alignment
\n");
-printf("
Show alignment between reference and query\n");
-printf("
\n");
+panelPrompt("collapseProtAli", "Show alignment between reference and query");
printf("
%s\n", info->prot_alignment);
printf("
\n");
// show inactivating mutations if required
printf("
\n");
printf("
List of inactivating mutations
\n");
-printf("
Show inactivating mutations\n");
-printf("
\n");
+panelPrompt("collapseMuts", "Show inactivating mutations");
printf("
\n"); // init table
printf("| Exon number | Codon number | Mutation class | Mutation | Treated as inactivating | Mutation ID | \n");
printf("
|---|
\n");
printf("%s\n", info->inact_mut_html_table);
printf("
\n");
printf("
\n\n");
// show exons data
printf("
\n");
printf("
Exon alignments
\n");
-printf("
Show exon sequences and features\n");
-printf("
\n");
+panelPrompt("collapseExons", "Show exon sequences and features");
printf("
%s
\n", info->exon_ali_html);
printf("
\n");
printf("
\n
\n");
printTrackHtml(tdb); // and do I need this?
}
void doHillerLabTOGAGene(char *database, struct trackDb *tdb, char *item, char *table_name)
/* Put up TOGA Gene track info. */
{
//int start = cartInt(cart, "o");
char headerTitle[512];
char suffix[512];
strcpy(suffix, table_name);
extractHLTOGAsuffix(suffix);
safef(headerTitle, sizeof(headerTitle), "%s", item);
genericHeader(tdb, headerTitle);
printf("
TOGA gene annotation
\n");
// htmlHorizontalLine();
if (startsWith("bigBed", tdb->type))
{
doHillerLabTOGAGeneBig(database, tdb, item, table_name);
return;
}
struct sqlConnection *conn = hAllocConn(database);
// define TOGA table names: initate with pre-defined prefixes
char togaDataTableName[256];
char togaNuclTableName[256];
char togaInactMutTableName[256];
strcpy(togaDataTableName, HLTOGA_DATA_PREFIX);
strcpy(togaNuclTableName, HLTOGA_NUCL_PREFIX);
strcpy(togaInactMutTableName, HLTOGA_INACT_PREFIX);
// add suffix
strcat(togaDataTableName, suffix);
strcat(togaNuclTableName, suffix);
strcat(togaInactMutTableName, suffix);
if (hTableExists(database, togaDataTableName))
{
printf("
Projection %s
\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("- 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("
\n
Feature 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("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("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("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("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("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("
\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");
+ panelPrompt("collapseGLP", "Show features used for transcript classification");
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("- Middle 80 percent of CDS present: %s
\n", YES_);
} else {
printf("- Middle 80 percent of CDS present: %s
\n", NO_);
}
printf("
\n
\n
\n");
printf("
Query protein sequence
");
- printf("
Show protein sequence of query\n");
- printf("
\n");
+ panelPrompt("collapseProt", "Show protein sequence of query");
printf("
{protein seq of the query without dashes or other things. Should end with *}\n");
printf("
\n
\n
\n");
// and show protein sequence
htmlHorizontalLine();
printf("
Protein sequence alignment
\n");
- printf("
Show alignment between reference and query\n");
- printf("
\n");
+ panelPrompt("collapseProtAli", "Show alignment between reference and query");
printf("
%s
\n", info->prot_alignment);
printf("
\n
\n");
// do not forget to free toga data struct
togaDataFree(&info);
} else {
// no data found, need to report this
printf("
No 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");
+ panelPrompt("collapseMuts", "Show inactivating mutations");
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("%s | \n", YES_);
} else {
printf("%s | \n", NO_);
}
printf("%s | \n", info->mut_id);
printf("
\n");
togaInactMutFree(&info);
}
sqlFreeResult(&sr);
printf("
\n");
printf("
\n
\n");
} else {
printf("
Sorry, cannot find TOGAInactMut table.\n");
}
// show exons data
htmlHorizontalLine();
printf("
Exon alignments
\n");
if (hTableExists(database, togaNuclTableName))
{
char query[256];
struct sqlResult *sr = NULL;
char **row;
- printf("
Show exon sequences and features\n");
- printf("
\n");
+ panelPrompt("collapseExons", "Show exon sequences and features");
sqlSafef(query, sizeof(query), "select * from %s where transcript='%s'", togaNuclTableName, item);
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
struct togaNucl *info = NULL;
info = togaNuclLoad(row);
printf("
Exon number: %s
\n", info->exon_num);
printf("Exon region: %s
\n", info->exon_region);
printf("Nucleotide percent identity: %s | BLOSUM: %s
\n", info->pid, info->blosum);
if (sameWord(info->gaps, ONE_)){
printf("Intersects assembly gaps: %s
\n", YES_);
} else {
printf("Intersects assembly gaps: %s
\n", NO_);
}
printf("Exon alignment class: %s
\n", info->ali_class);
if (sameWord(info->in_exp_region, ONE_)){
printf("Detected within expected region (%s): %s
\n", info->exp_region, YES_);
} else {
printf("Detected within expected region (%s): %s
\n", info->exp_region, NO_);
}
// printf("Expected region: %s
\n", info->exp_region);
printf("
\n");
printf("Sequence alignment between reference and query exon:
\n");
printf("%s
\n", info->alignment);
togaNuclFree(&info);
}
sqlFreeResult(&sr);
printf("\n
\n");
} else {
printf("
Sorry, cannot find TOGANucl table.\n");
}
htmlHorizontalLine();
// TODO: check whether I need this
printf("%s", hgTracksPathAndSettings());
printTrackHtml(tdb); // and do I need this?
hFreeConn(&conn);
}
%s: