clusterTraces: cluster traces in space (and/or time) using DBscan

In jgodet/trackR: R tools and functions to compute jump analysis, describe and infer from single-molecule tracking data

Description

cluster traces in space (and/or time) using DBscan

Usage

clusterTraces(dtf, coord = c("x", "y"), eps, minPts, method, ...)

Arguments

dtf	data frame with coordinates and traces ID
coord	coordinates dataframe (x,y) or (x,y,t)
eps	Reachability distance, see Ester et al. (1996).
minPts	Reachability minimum no. of points, see Ester et al. (1996).
method	"dist" treats data as distance matrix (relatively fast but memory expensive), "raw" treats data as raw data and avoids calculating a distance matrix (saves memory but may be slow), "hybrid" expects also raw data, but calculates partial distance matrices (very fast with moderate memory requirements).

Author(s)

JuG

Examples

xmlPath <- "//Users/jgodet/Seafile/MaBibliotheque/Hanna/1_3_MMStack_Pos0.ome.xml"
data <- readTrackMateXML(XMLpath = xmlPath)
data$jump<-jump(data, spaceRes=1)
bacteria <- clusterTraces(dtf = data, eps = .2,minPts = 100)
table(bacteria)
bact1 <- data[which(bacteria==7),]
summary(bact1)
bact1 %>% select(x,y) %>% plot(., asp=1)
drawRod(data = getContour(bact1),col='green')
drawRod(data = getContour(cleanNearest(data = bact1,k = 3)),col='blue')

jgodet/trackR documentation built on May 24, 2020, 2:21 p.m.