igPCA 1.0.0

Integrative Generalized Principle Analysis

Intergative Generalized Principle Componenet Analysis (igPCA)

Intergative Generalized Principle Componenet Analysis (igPCA) is a framework for joint dimensionality reduction of double-structure data. The algorithm details wiil be available soon in out manuscript: Xie X and Ma J (2023). * Structured dimensionality reduction for multi-view microbiome data*

igPCA is a python implementation of the proposed framework.

Installation

$ pip install igPCA

Usage

igPCA can be used to perform joint dimensinality reduction for two dataset X1 and X2 as follows:

from .igPCA import igPCA
import matplotlib.pyplot as plt

model = igPCA(X1, X2, H, Q1, Q2, r1, r2)
model.fit(r0 = r0)

In this simple example, H, Q1 and Q2 are kernel matrices characterzing X1 and X2. The total rank for X1 and X2 are r1 and r2, respectively. r0 is the joint rank between X1 and X2.