runDCCA: Perform canonical correlation analysis (CCA) on two dimension

In jinzhuangdou/DCCA: Tools for single cell multi-omics analysis \

Description

Assuming there are two matrix X (M genes by K cells) and Y (N loci by L cells), we want to find the cell correspondences and correlated gene/loci module between them. Traditional CCA could not handle this case because it requires that there should be at least one dimension shared by two datasets. The DCCA function aims to finding one transition matrix Z (M genes by L cells) to bridge X and Y. More details can be seen in xx

Usage

runDCCA(
  X = NULL,
  Z0 = NULL,
  Y = NULL,
  num.D = 10,
  output = NULL,
  num.E = 10,
  num.iteration = 100,
  tolerance = 0.01,
  save = FALSE
)

Arguments

X	First matrix (M genes by K cells)
Z0	Transition matrix (M genes by L cells)
Y	Second matrix (N loci by L cells)
num.D	Number of canonical vectors to calculate for pair (X, Z) [default 10]
num.E	Number of canonical vectors to calculate for pair (Z, Y) [default 10]
num.iteration	Maximal number of iteration [default 100]
tolerance	Relative change ratio for ||Z||F during iteration [default 0.01]
save	Save the temporay files [default FALSE]

Value

Returns the object with list:

* 'Z' - contains the estimated transition matrix (M genes by L cells) * 'u' - contains the canonical correlation vectors for X (K cells by num.D factor) * 'r' - contains the canonical correlation vectors for Z (L cells by num.D factor) * 's' - contains the canonical correlation vectors for Z (M genes by num.E factor) * 'v' - contains the canonical correlation vectors for Z (N loci by num.E factor)

Author(s)

Jinzhuang Dou, [e-mail]

See Also

xx xx

Examples

runDCCA(X=X, Y=Y, Z0 = Z0, num.D = 6, num.E = 6)

jinzhuangdou/DCCA documentation built on June 29, 2020, 12:54 a.m.