clusterSVs: Cluster SVs based on overlap/similarity

In jmonlong/sveval: SV evaluation

Cluster SVs based on overlap/similarity

Description

Cluster SVs based on overlap/similarity

Usage

clusterSVs(
  svs.gr,
  min.rol = 0.8,
  max.ins.dist = 20,
  range.seq.comp = FALSE,
  ins.seq.comp = FALSE,
  simprep = NULL,
  nb.cores = 1,
  batch.maxsize = 5000,
  log.level = c("CRITICAL", "WARNING", "INFO")
)

Arguments

svs.gr	A GRanges with SVs (for example read by readSVvcf or readSVvcf.multisamps)
min.rol	minimum reciprocal overlap for deletions and other "ranges" SVs. Default is 0.1
max.ins.dist	maximum distance for insertions to be clustered.
range.seq.comp	compare sequence instead of overlapping deletions/inversion/etc. Default is FALSE.
ins.seq.comp	compare sequence instead of insertion sizes. Default is FALSE.
simprep	optional simple repeat annotation. Default is NULL. If non-NULL, GRanges to be used to
nb.cores	number of processors to use. Default is 1.
batch.maxsize	batch size to aim. To reduce memory usage, see Details. Default is 5000.
log.level	the level of information in the log. Default is "CRITICAL" (basically no log).

Details

SVs are overlapped with each other. A graph is then built where nodes (SVs) are connected is they overlap/match. A cluster is a component in this graph.

To reduce the memory usage, the SVs are first grossly clustered into batches. The actual clustering (and graph construction) is then performed separately for each batch, potentially in parallel.

Value

the svs.gr object annotated with two new columns:

svsite	the ID of the cluster (or SV site)
clique	is this cluster a clique, i.e. all SVs overlapping/matching all other SVs in the cluster

Author(s)

Jean Monlong

Examples

## Not run: 

svs = readSVvcf('svs.vcf.gz', keep.ids=TRUE)
svs = clusterSVs(svs)


## End(Not run)

jmonlong/sveval documentation built on July 31, 2023, 7:50 p.m.