pcaplot: 2d and 3d sample relationship plot
In Liuy12/MBDDiff: MBDDiff for processing MBDcap-seq datasets

Description Usage Arguments Details Value Author(s) References See Also Examples

Generate 2d or 3d sample relationship plot based on principal component analysis, multidimensional scaling, etc.

pcaplot(x, subset = NULL, cv.Th = 0.1, var.Th = 0, mean.Th = 0, standardize = TRUE, method = c("cluster", "mds", "pca"), dimension = c(1, 2, 3), color = "black", princurve = FALSE, lwd = 1, starts = NULL, col.curve = "red", text = TRUE, main = NULL, psi = 4, type = "p", ...)

`x`	A matrix of numeric values
`subset`	A numeric value indicating the number of genes that is randomly selected for pca analysis. Default to NULL, which means no subsetting procedure will be done
`cv.Th`	If subset = NULL, a numeric value indicating threshold of coeffcient of variation in selecting genes
`var.Th`	A numeric value indicating threshold of variation in selecting genes. This is only used when subset and cv.Th are both set to NULL
`mean.Th`	Similar to var.Th, a numeric value indicating threshold of mean value in selecting genes. This is only used when subset and cv.Th are both set to NULL
`standardize`	Whether to standardize samples so that each sample has mean 0 and variance 1. Default to TRUE
`method`	One of "cluster", "mds", "pca". Please refer to details section
`dimension`	Numeric vector indicating the number of dimensions you would like to generate the figure
`color`	Color for points when 'mds' or 'pca' is chosen as method
`princurve`	Logical value indicating whether to generate a principal curve. Please refer to details
`lwd`	The line width for principal curve
`starts`	Providing the starting point for principal curve. Please refer to details
`col.curve`	The color of principal curve
`text`	Logical value indicating whether text is added as label to the figure
`main`	Main title for the figure
`psi`	Integer value indicating point size
`type`	For the default method, a single character indicating the type of item to plot. Supported types are: 'p' for points, 's' for spheres, 'l' for lines, 'h' for line segments from z = 0, and 'n' for nothing.
`...`	Further arguments will be ignored

If method = 'cluster', hclust is used; if method = 'mds', cmdscale is used; if method = 'pca', prcomp is used.

If princurve is set to TURE, then fits a principal curve which describes a smooth curve that passes through the middle of the data x in an orthogonal sense. This curve is a nonparametric generalization of a linear principal component. For details of principal curve, please refer to principal.curve. When princurve is set to TRUE, you need to provide a starting point for principal curve as starts argument. starts is basically a logical vector of the same length as number of samples, that tells you which sample will be used as starting point.

If method = 'cluster': a 'hclust' object If method = 'mds' or 'pca', a data.frame containing user specified number of principal components.

Yuanhang Liu

https://github.com/Liuy12/MBDDiff

cmdscale, hclust, prcomp, principal.curve

    ## Not run: 
        data(PromoterCount)
        Condition <- c(rep('C1', 3), rep('C2', 3))
        TestStat <- MBDDiff(Promoter, Background, Condition)
        MBD <- TestStat[[1]]
        Norm_count <- counts(MBD, normalized = TRUE)
        pcaplot(Norm_count, cv.Th = 0.1, method = 'pca', dimension = c(1,2,3), princurve = TRUE, starts = c(1,1,1,0,0,0)) 
        
## End(Not run)