Learning Massive-scale Partial Correlation Networks in Clinical Multi-omics Studies with HP-ACCORD

TL;DR

This paper presents a scalable, pseudolikelihood-based graphical model framework for estimating partial correlation networks from ultra-high-dimensional multi-omics data, enabling biological insights with improved specificity.

Contribution

It introduces a novel reparameterization and optimization approach that allows for fast, scalable estimation of sparse precision matrices in high-dimensional settings.

Findings

Successfully estimated networks with up to one million variables.

Demonstrated superior specificity in identifying key biological regulators.

Validated the method on liver cancer multi-omics data.

Abstract

Graphical model estimation from multi-omics data requires a balance between statistical estimation performance and computational scalability. We introduce a novel pseudolikelihood-based graphical model framework that reparameterizes the target precision matrix while preserving the sparsity pattern and estimates it by minimizing an $\ell_1$-penalized empirical risk based on a new loss function. The proposed estimator maintains estimation and selection consistency in various metrics under high-dimensional assumptions. The associated optimization problem allows for a provably fast computation algorithm using a novel operator-splitting approach and communication-avoiding distributed matrix multiplication. A high-performance computing implementation of our framework was tested using simulated data with up to one million variables, demonstrating complex dependency structures similar to those found in biological networks. Leveraging this scalability, we estimated a partial correlation network from a dual-omic liver cancer data set. The co-expression network estimated from the ultrahigh-dimensional data demonstrated superior specificity in prioritizing key transcription factors and co-activators by excluding the impact of epigenetic regulation, thereby highlighting the value of computational scalability in multi-omic data analysis.