Question: What is the main difference between nclist and intervaltree algorithm in GenomicRanges packages ?
0
3.8 years ago by
Chicago, IL, USA
Jurat Shahidin70 wrote:

In new version of bioconductor, I have used GenomicRanges packages, but intervaltree is defunct. I believed that intervaltree algorithm is used for finding overlapped regions from different genomic interval. What is the main different between intervaltree and nclist algorithm for finding overlapped regions in R ? I am bit of confused. I would be grateful if someone gives me comprehend explanation for it ? FYI, finding overlapped regions in parallel, using intervaltree is only way to solve the problem ? Thanks a lot the benefit of this community.

genomeintervals nclist • 710 views
modified 3.8 years ago by Hervé Pagès ♦♦ 14k • written 3.8 years ago by Jurat Shahidin70
Answer: What is the main difference between nclist and intervaltree algorithm in Genomic
3
3.8 years ago by
Hervé Pagès ♦♦ 14k
United States
Hervé Pagès ♦♦ 14k wrote:

Short answer: They return basically the same thing but the new algo (nclist) is more efficient.

Long answer: Here is what the documentation says about the algorithm argument (documented in ?IRanges::findOverlaps):

• algorithm: Can be "nclist" (the default) or "intervaltree".  This argument was added in BioC 3.1 to facilitate the transition between the new findOverlaps/countOverlaps implementation based on Nested Containment Lists and the old implementation based on Interval Trees.  See ?NCList and ?IntervalTree for more information about these implementations. Note that the old implementation is defunct starting with BioC 3.2. The algorithm argument will be removed in BioC 3.3.

The man page for NCList objects (?NCList) in particular contains information about the differences between the 2 algorithms.

H.