be36584718d57384b01657483422f51adf29f0bb
max
Sat Jul 25 18:22:02 2026 -0700
vcfTrack: add minimum alternate allele count (minAc) filter, refs #37927
When a VCF INFO column includes AC, offer a 'minimum allele count' filter in the
track UI (e.g. 2 to hide singletons) and drop records whose largest alternate
allele count is below the threshold.
diff --git src/hg/lib/vcfUi.c src/hg/lib/vcfUi.c
index bb3a5188e0c..7ba16bd1580 100644
--- src/hg/lib/vcfUi.c
+++ src/hg/lib/vcfUi.c
@@ -430,30 +430,44 @@
static void vcfCfgMinAlleleFreq(struct cart *cart, struct trackDb *tdb, struct vcfFile *vcff,
char *name, boolean parentLevel)
/* Show input for minimum allele frequency, if we can extract it from the VCF INFO column. */
{
printf("Minimum minor allele frequency (if INFO column includes AF or AC+AN):\n");
double cartMinFreq = cartOrTdbDouble(cart, tdb, VCF_MIN_ALLELE_FREQ_VAR,
VCF_DEFAULT_MIN_ALLELE_FREQ);
char varName[1024];
safef(varName, sizeof(varName), "%s." VCF_MIN_ALLELE_FREQ_VAR, name);
cgiMakeDoubleVarInRange(varName, cartMinFreq, "minor allele frequency between 0.0 and 0.5", 5,
"0.0", "0.5");
puts("
");
}
+static void vcfCfgMinAc(struct cart *cart, struct trackDb *tdb, struct vcfFile *vcff,
+ char *name, boolean parentLevel)
+/* Show input for minimum alternate allele count, if INFO column includes AC. This lets
+ * the user filter on a whole-number count (e.g. 2 to hide singletons) instead of having to
+ * enter a tiny frequency cutoff. */
+{
+printf("Minimum allele count (if INFO column includes AC), e.g. 2 to hide singletons:\n");
+int cartMinAc = cartOrTdbInt(cart, tdb, VCF_MIN_AC_VAR, VCF_DEFAULT_MIN_AC);
+char varName[1024];
+safef(varName, sizeof(varName), "%s." VCF_MIN_AC_VAR, name);
+cgiMakeIntVarWithMin(varName, cartMinAc, "minimum allele count", 5, 0);
+puts("
");
+}
+
static char *getChildSample(struct trackDb *tdb)
/* Return just the VCF sample name of the phased trio child setting */
{
char *childSampleMaybeAlias = cloneString(trackDbLocalSetting(tdb, VCF_PHASED_CHILD_SAMPLE_SETTING));
char *pt = strchr(childSampleMaybeAlias, '|');
if (pt != NULL)
*pt = '\0';
return childSampleMaybeAlias;
}
static struct slPair *vcfPhasedGetSamplesFromTdb(struct trackDb *tdb, boolean hideOtherSamples)
/* Get the different VCF Phased Trio setings out of trackDb onto a list */
{
// cloneString here because we will be munging the result if there are alternate labels
char *childSampleMaybeAlias = cloneString(trackDbLocalSetting(tdb, VCF_PHASED_CHILD_SAMPLE_SETTING));
@@ -688,48 +702,51 @@
printInfoIcon(phasedInfoText);
}
void vcfCfgUi(struct cart *cart, struct trackDb *tdb, char *name, char *title, boolean boxed)
/* VCF: Variant Call Format. redmine #3710 */
{
boxed = cfgBeginBoxAndTitle(tdb, boxed, title);
printf("
", boxed ? " width='100%'" : "");
struct vcfFile *vcff = vcfHopefullyOpenHeader(cart, tdb);
if (vcff != NULL)
{
boolean parentLevel = isNameAtParentLevel(tdb, name);
boolean doVcfFilterUi = cartOrTdbBoolean(cart, tdb, VCF_DO_FILTER_UI, TRUE);
boolean doVcfQualUi = cartOrTdbBoolean(cart, tdb, VCF_DO_QUAL_UI, TRUE);
boolean doVcfMafUi = cartOrTdbBoolean(cart, tdb, VCF_DO_MAF_UI, TRUE);
+ boolean doVcfMinAcUi = cartOrTdbBoolean(cart, tdb, VCF_DO_MIN_AC_UI, TRUE);
if (vcff->genotypeCount > 1 && !sameString(tdb->type, "vcfPhasedTrio"))
{
vcfCfgHapCluster(cart, tdb, vcff, name, parentLevel);
}
if (sameString(tdb->type, "vcfPhasedTrio"))
{
vcfCfgPhasedTrioUi(cart, tdb, vcff, name, parentLevel);
}
if (differentString(tdb->track,"evsEsp6500") && (doVcfFilterUi || doVcfQualUi))
{
puts("Filters"); if (doVcfQualUi) vcfCfgMinQual(cart, tdb, vcff, name, parentLevel); if (doVcfFilterUi) vcfCfgFilterColumn(cart, tdb, vcff, name, parentLevel); } if (doVcfMafUi) vcfCfgMinAlleleFreq(cart, tdb, vcff, name, parentLevel); + if (doVcfMinAcUi) + vcfCfgMinAc(cart, tdb, vcff, name, parentLevel); } else { printf("Sorry, couldn't access VCF file.\n"); } puts(" | ");
if (boxed && fileExists(hHelpFile("hgVcfTrackHelp")))
printf("VCF configuration help | "); printf("