lshift2: lshift2
In deformetrics/deformetrics: R functions for shape analysis

Description Usage Arguments Value Author(s) References Examples

This function performs the Euclidean Parallel Transport, i.e. the Linear Shift strategy, described in Piras et al (2016). It applies a series of deformations occurring in different groups to a Common Reference by properly managing rotations.

1 2	lshift2(array, factor, CR = NULL, locs = NULL, CSinit = F, scale = F, reflect = F, reorder = F, sc = F)

`array`	numeric: an array kxmxn of landmark coordinates
`factor`	character vector: a factor with group affiliation
`CR`	logical: if NULLL CR is estimated as the consensus of the array via Generalized Procrustes Analysis (GPA). Otherwise the specified CR is used.
`reflect`	logical: “reflect” argument in procSym() from package “Morpho”
`reorder`	logical: If TRUE the factor is reordered
`sc`	logical: This is not supposed to be called by common user
`locs=NULL:`	if NULL the local references are estimated via separate per-group GPAs. Oterwise the specified locs are used.
`CSinit=F`	logical: if TRUE scaling at unit size is performed.
`scale=F`	logical “scale” argument in procSym() from package “Morpho”

transported list: a list of length 1 containing the array with the transported shapes.

Paolo Piras

Piras P., Teresi L., Traversetti L., Varano V, Gabriele S., Kotsakis T., Raia P., Puddu P.E., Scalici M. (2016). The conceptual framework of ontogenetic trajectories: Parallel Transport allows the recognition and visualization of pure deformation patterns. Evolution and Development 18: 182-200. doi: 10.1111/ede.12186

## Not run:  
data(case1)
group<-factor(rep(1:5,each=21))
linksdors<-conslinks(21)
library(gdata)
lscase1<-lshift2(case1,group,CR=procSym(case1[,,firstsfac(group)],CSinit=F,reflect=F,scale=F,pcAlign=F)$mshape,locs=case1[,,firstsfac(group)])
plottraj(procSym(lscase1$transported,CSinit=T)$PCscores[,1:2],group,asp=1) #### very similar to Levi Civita

## End(Not run)