analysis

Multi-Omics Analysis

Integration methods for combining genomics, transcriptomics, proteomics, and epigenomics data layers using joint factorization and correlation approaches.

Contents

File	Purpose
integration.py	Multi-omics data container, joint PCA, NMF, CCA, and data converters

Key Classes and Functions

Symbol	Description
MultiOmicsData	Container for multiple omics layers with sample alignment
integrate_omics_data()	Combine multiple omics DataFrames into unified representation
joint_pca()	Joint PCA across concatenated omics layers
joint_nmf()	Non-negative matrix factorization across omics layers
canonical_correlation()	Canonical correlation analysis between two omics types
from_dna_variants()	Convert VCF data to integration-ready format
from_rna_expression()	Convert expression matrix to integration-ready format
from_protein_abundance()	Convert protein quantification to integration-ready format
from_epigenome_data()	Convert methylation or ChIP data to integration-ready format

Usage

from metainformant.multiomics.analysis.integration import (
    MultiOmicsData,
    integrate_omics_data,
    joint_pca,
)

multi = MultiOmicsData(layers={"rna": rna_df, "protein": prot_df})
integrated = integrate_omics_data([rna_df, prot_df])
components = joint_pca(integrated, n_components=10)