countOvpPACds: Count overlapping between PACdatasets

In BMILAB/movAPA: movAPA: Modeling and Visualization of Dynamics of Alternative PolyAdenylation

Count overlapping between PACdatasets

Description

countOvpPACds counts overlapping between PACdatasets.

Usage

countOvpPACds(
  pacds1,
  pacds2,
  mode1 = "point",
  mode2 = "point",
  d = 0,
  pats1 = 1,
  pats2 = 1
)

Arguments

pacds1	a PACdataset.
pacds2	another PACdataset.
mode1	use coord or UPA_start/UPA_end for pacds1, default is point (using coord). If mode1=region, then use UPA_start/UPA_end.
mode2	same as mode1 except for pacds2.
d	distance to count overlapping, default is 0 (no gap between qry and sbj).
pats1	cutoffs to filter pacds1 by total tag number of all samples. If pats1<=1, then do not filter. Multiple cutoffs can be set, like pats1=c(10, 20, 30).
pats2	same as pats1 but to filter pacds2.

Details

This function is used to compare PA coordinates of two PACdataset objects.

Value

A dataframe with 8 columns: PAT1, PAT2, Total1, Total2, Ovp1, Ovp2, Ovp1Pct, Ovp2Pct. Row number is the combination number of pats1 and pats2.

Examples

data(PACds)
pacds1=PACds
pacds2=PACds
pacds2@anno$coord=pacds2@anno$coord+sample(-10:10, 1)
## Count exactly the same positions between two pacds without filtering
countOvpPACds(pacds1, pacds2, d=0)
## count overlapping with different cutoffs
countOvpPACds(pacds1, pacds2, d=50, pats1=1, pats2=seq(0,20,10))
## count overlapping with different cutoffs, comparing ds1's coord ~ ds2's region
countOvpPACds(pacds1, pacds2, mode1='point', mode2='region', d=50, pats1=1, pats2=seq(0,20,10))

BMILAB/movAPA documentation built on Jan. 3, 2024, 11:09 p.m.