Article

Protein-protein interaction priors shape biologically coherent latent spaces for causally concordant cross-omic translation

2025-10-14

Abstract excerpt

<h4>ABSTRACT</h4> Deep learning models routinely compress omics into low-dimensional codes, yet many equally accurate embeddings fail to reflect how cells are wired, which limits explanation and causal reasoning. We present a simple, architecture-agnostic approach to make latent spaces biologically legible: a protein-protein interaction (PPI) prior that softly steers autoencoder units to recruit genes that are pr...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 34e35319-02bc-516f-989e-9dcc45b308c3
DOI: 10.1101/2025.10.13.681970

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Protein-protein interaction priors shape biologically coherent latent spaces for causally concordant cross-omic translationDOI 10.1101/2025.10.13.681970Read

Data-driven representations using deep network-coherent methylation autoencoders to identify robust disease and risk factor signaturesSemantic proximity Read

PIMO: Pathway-based Interpretable Multi-Omics interactions for multi-omics integrationSemantic proximity Read

CoxFormer enables spatial omics inference with multimodal generative modelingSemantic proximity Read

From Black Box to Biological Insight: AttentioFuse Unlocks Multi-Omics Dynamics in Lung CancerSemantic proximity Read

Multi-omics integration and regulatory inference for unpaired single-cell data with a graph-linked unified embedding frameworkSemantic proximity Read

Scalable Non-negative Matrix Factorization of the Human Cell Census Reveals Interpretable Transcriptional ProgramsSemantic proximity Read

Learning the spatial cell-cell communication network to decode multi-channel signaling and predict network-hub vulnerabilities with MOSANICSemantic proximity Read

System-level health profiling from blood DNA methylation with explainable deep learningSemantic proximity Read

GAE-Δ: A Graph-Learning Framework for Gene Network Rewiring and Clinical Outcome Prediction from Multi-Omics DataSemantic proximity Read

DeepSpot-M: a multimodal foundation model for transcriptome-wide virtual spatial transcriptomics from histologySemantic proximity Read

CAPTAIN: A multimodal foundation model pretrained on co-assayed single-cell RNA and proteinSemantic proximity Read

Revealing the impact of genomic alterations on cancer cell signaling with a partially transparent deep learning modelSemantic proximity Read