This track combines the structural-variant (SV) callsets from the individual subtracks of the Long-read SVs supertrack into a single, position-merged overview. Each item is an SV locus seen in one or more of the contributing long-read databases. For every merged locus the track records which databases report it, the summed allele count across those databases, and the range of allele frequencies observed, making it useful for quickly seeing how widely an SV has been reported across cohorts.
This is a summary view. For cohort-specific genotypes, per-population allele frequencies, and dataset-specific annotations, use the individual subtracks of the supertrack. The merge includes the released long-read callsets only; preliminary or unpublished subtracks (e.g. the Kim PD brain, 1000 Genomes linear, and HPRC Jasmine sets) are not part of this merged track.
Items are colored by SV type, matching the individual subtracks:
The mouseover shows the variant name, SV type, reference and insertion lengths, the list of contributing source databases, the allele-frequency range across those databases, and the total allele count. Filters are available for the source database, SV type, SV length, insertion length, total allele count, minimum and maximum allele frequency, and the number of source databases reporting each locus. The detail page lists the per-database allele counts.
The merged track is built by the lrSvMergeAll.py script, which reads the bigBed of each contributing subtrack (configured in databases.tsv) and groups records that share an identical (chromosome, start, end) position and SV type. For each merged locus the script records the set of contributing databases (sources), the number of those databases (sourceCount), the sum of their allele counts (AC), and the minimum and maximum allele frequency across databases that report one (minAF, maxAF). The per-database allele counts are carried as additional columns.
The step-by-step build commands are recorded in the UCSC makeDoc for this track collection: doc/hg38/lrSv.txt. The merge script and autoSql schema live in -makeDb/scripts/lrSv. +makeDb/scripts/lrSv, and the track configuration is in trackDb/human/lrSvAll.ra.
The data can be explored interactively with the Table Browser or the Data Integrator, and accessed programmatically through our API, track=lrSvAll.
The bigBed is available from our download server as lrSvAll.bb. Example: bigBedToBed http://hgdownload.soe.ucsc.edu/gbdb/$db/lrSv/lrSvAll.bb -chrom=chr21 -start=0 -end=100000000 stdout.
This merged view is derived entirely from the contributing long-read SV callsets; please see the individual subtrack description pages for the data producers and citations for each cohort.