IPF_alg: Iterative proportional fitting algorithm

In AnnikaGable/chromium: Analyze and visualize high throughput chromatin interaction data, such as Hi-C or ChIA-PET

Description

Iterative proportional fitting (IPF) is used to normalize a sparse matrix.

Usage

IPF_alg(symLFM, numberOfIterations)

Arguments

symLFM	A symmetric sparse ligation frequency matrix (LFM) with binned or unbinned data.
numberOfIterations	The number of iterations for normalization. Usually, 10 should be more than enough.

Value

A list containing the normalized, symmetric matrix normLFM, the dataframes with the row sums rowBiases and the column sums colBiases of every iteration, and a vector with the loss function lf for every iteration lossFunction.

Examples

M <- matrix(c(3,0,1,0,7,9,1,9,7),3,3)
M <- Matrix::Matrix(M, sparse = TRUE)
ipf <- chromium:::IPF_alg(M,10)

AnnikaGable/chromium documentation built on May 5, 2019, 6:04 a.m.