Saved in:
Bibliographic Details
Main Authors: Piao, Yinhua, Kim, Hyomin, Kim, Seonghwan, Oh, Yunhak, Jeon, Junhyeok, Hwang, Sang-Yeon, Lim, Jaechang, Kim, Woo Youn, Park, Chanyoung, Ahn, Sungsoo
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.18885
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917286453968896
author Piao, Yinhua
Kim, Hyomin
Kim, Seonghwan
Oh, Yunhak
Jeon, Junhyeok
Hwang, Sang-Yeon
Lim, Jaechang
Kim, Woo Youn
Park, Chanyoung
Ahn, Sungsoo
author_facet Piao, Yinhua
Kim, Hyomin
Kim, Seonghwan
Oh, Yunhak
Jeon, Junhyeok
Hwang, Sang-Yeon
Lim, Jaechang
Kim, Woo Youn
Park, Chanyoung
Ahn, Sungsoo
contents Predicting high-dimensional transcriptional responses to genetic perturbations is challenging due to severe experimental noise and sparse gene-level effects. Existing methods often suffer from mean collapse, where high correlation is achieved by predicting global average expression rather than perturbation-specific responses, leading to many false positives and limited biological interpretability. Recent approaches incorporate biological knowledge graphs into perturbation models, but these graphs are typically treated as dense and static, which can propagate noise and obscure true perturbation signals. We propose AdaPert, a perturbation-conditioned framework that addresses mean collapse by explicitly modeling sparsity and biological structure. AdaPert learns perturbation-specific subgraphs from biological knowledge graphs and applies adaptive learning to separate true signals from noise. Across multiple genetic perturbation benchmarks, AdaPert consistently outperforms existing baselines and achieves substantial improvements on DEG-aware evaluation metrics, indicating more accurate recovery of perturbation-specific transcriptional changes.
format Preprint
id arxiv_https___arxiv_org_abs_2602_18885
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning Adaptive Perturbation-Conditioned Contexts for Robust Transcriptional Response Prediction
Piao, Yinhua
Kim, Hyomin
Kim, Seonghwan
Oh, Yunhak
Jeon, Junhyeok
Hwang, Sang-Yeon
Lim, Jaechang
Kim, Woo Youn
Park, Chanyoung
Ahn, Sungsoo
Computational Engineering, Finance, and Science
Predicting high-dimensional transcriptional responses to genetic perturbations is challenging due to severe experimental noise and sparse gene-level effects. Existing methods often suffer from mean collapse, where high correlation is achieved by predicting global average expression rather than perturbation-specific responses, leading to many false positives and limited biological interpretability. Recent approaches incorporate biological knowledge graphs into perturbation models, but these graphs are typically treated as dense and static, which can propagate noise and obscure true perturbation signals. We propose AdaPert, a perturbation-conditioned framework that addresses mean collapse by explicitly modeling sparsity and biological structure. AdaPert learns perturbation-specific subgraphs from biological knowledge graphs and applies adaptive learning to separate true signals from noise. Across multiple genetic perturbation benchmarks, AdaPert consistently outperforms existing baselines and achieves substantial improvements on DEG-aware evaluation metrics, indicating more accurate recovery of perturbation-specific transcriptional changes.
title Learning Adaptive Perturbation-Conditioned Contexts for Robust Transcriptional Response Prediction
topic Computational Engineering, Finance, and Science
url https://arxiv.org/abs/2602.18885