palmer: PALMER: A Constrained Biclustering Algorithm to Improve...

In elflini/palmer: PALMER: A Constrained Biclustering Algorithm to Improve Pathway Annotation Based on the Biomedical Literature Mining

Description

Fit PALMER model

Usage

palmer(X, path = NULL, K = 3, L = 3, B = 1000)

Arguments

X	Matrix of binary values with n samples by p variables
path	List. Gene list for the prior information (default: NULL)
K	The number of gene clusters (default: 3)
L	The number of GO term clusters (default: 3)
B	The number of bootstrappings (default: 1000)

Details

palmer fits the PALMER model. It requires binary matrix to be clusterd and list of gene set for the constraints.

Value

Construct palmer class object.

Author(s)

Jin Hyun Nam

References

Nam JH, Couch D, Silveira W.A, Yu Z and Chung D (2019) ”PALMER: A constrained biclustering Algorithm to improve pathway annotation based on the biomedical literature mining”.

Examples

# Loading data

data(sdata)
data(pathway)

# Fit PALMER

fit.palmer <- palmer(X=sdata,path=pathway,K=2,L=3,B=100)
fit.palmer

# Extract Gene and GO term cluster and its probability

predict(fit.palmer)

# Visualize the clustering analysis result

plot(fit.palmer)

elflini/palmer documentation built on July 30, 2020, 7:05 p.m.