analiza aiqbench
This research introduces PerturbEnergy, a conditional energy-based prior designed to isolate perturbation-specific signals within complex cellular systems for drug discovery applications. By integrating into autoencoder architectures, the method explicitly tilts latent spaces toward perturbation-relevant regions, achieving significant improvements in recognizing gene expression responses across multiple benchmarks. The work advances machine learning techniques for biological data modeling, offering a robust framework for predicting cellular reactions to therapeutic interventions.
Learning Perturbation-Specific Representations through Conditional Energy Prior for Cellular Response Prediction
arXiv:2610.18456v1 Announce Type: cross Abstract: AI-based cellular perturbation prediction is essential to accelerate drug discovery and therapeutic identification. However, perturbation-specific signals are generally obscured by cellular complexity, making it difficult for existing methods to correctly capture perturbation effects. To address this problem, we propose PerturbEnergy, which introduces a conditional energy-based prior over perturbation representations. It assigns lower energy (higher probability) to representations compatible with the given perturbation and higher energy to others, thereby explicitly tilting the latent space toward perturbation-preferred regions. Besides, PerturbEnergy can be integrated into autoencoder architectures as we find its Evidence Lower Bound (ELBO) is equivalent to the ELBO of the complete autoencoder. We further prove that optimizing PerturbEnergy increases the lower bound on the conditional mutual information between representation and perturbation. Extensive experiments on four benchmark datasets demonstrate the effectiveness of our PerturbEnergy: it improves perturbation recognition especially in latent space with improvement of up to 83.2\%; It also achieves state-of-the-art performance for predicting top-50/100/200 differentially expressed gene(DEG) response and remaining competitive on highly variable gene(HVG) prediction. The code is provided: https://github.com/BlueWhaleLab/PerturbEnergy
Abstract:AI-based cellular perturbation prediction is essential to accelerate drug discovery and therapeutic identification. However, perturbation-specific signals are generally obscured by cellular complexity, making it difficult for existing methods to correctly capture perturbation effects. To address this problem, we propose PerturbEnergy, which introduces a conditional energy-based prior over perturbation representations. It assigns lower energy (higher probability) to representations compatible with the given perturbation and higher energy to others, thereby explicitly tilting the latent space toward perturbation-preferred regions. Besides, PerturbEnergy can be integrated into autoencoder architectures as we find its Evidence Lower Bound (ELBO) is equivalent to the ELBO of the complete autoencoder. We further prove that optimizing PerturbEnergy increases the lower bound on the conditional mutual information between representation and perturbation. Extensive experiments on four benchmark datasets demonstrate the effectiveness of our PerturbEnergy: it improves perturbation recognition especially in latent space with improvement of up to 83.2\%; It also achieves state-of-the-art performance for predicting top-50/100/200 differentially expressed gene(DEG) response and remaining competitive on highly variable gene(HVG) prediction. The code is provided: this https URL
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2610.18456 [cs.LG]
(or arXiv:2610.18456v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.18456 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 270 cuvinte
Subiecte conexe
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · 8 oct. 2026
Synthesis Superintelligence: from Semiconductors to Superconductors — Periodic Labs’ Liam Fedus and Ekin Dogus Cubuk
Latent Space · 8 oct. 2026
Finite-Time Node Separation in Recurrent Graph Neural Networks with Persistent Gaussian Perturbations
arXiv · 16 sept. 2026
Multimodal LLMs Can Learn to Read Brain Signals: A Vision--Language Model for Unified Multi-Task EEG Decoding
arXiv · 8 oct. 2026