Saved in:
Bibliographic Details
Main Authors: Yan, Xichen, Zang, Zelin, Chi, Changxi, Zhou, Jingbo, Yu, Chang, Wu, Jinlin, Cheng, Shenghui, Yang, Fuji, Luo, Jiebo, Lei, Zhen, Li, Stan Z.
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.18576
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916023155818496
author Yan, Xichen
Zang, Zelin
Chi, Changxi
Zhou, Jingbo
Yu, Chang
Wu, Jinlin
Cheng, Shenghui
Yang, Fuji
Luo, Jiebo
Lei, Zhen
Li, Stan Z.
author_facet Yan, Xichen
Zang, Zelin
Chi, Changxi
Zhou, Jingbo
Yu, Chang
Wu, Jinlin
Cheng, Shenghui
Yang, Fuji
Luo, Jiebo
Lei, Zhen
Li, Stan Z.
contents A critical challenge in single-cell RNA sequencing (scRNA-seq) integration is resolving the tension between eliminating batch effects and maintaining biological fidelity. While recent evidence indicates that batch effects manifest heterogeneously across genes, most existing methods process the transcriptome uniformly, frequently resulting in over-correction and loss of subtle biological signals. To address this, we present scHelix, a dataset-adaptive framework that fundamentally changes how features are processed by explicitly partitioning genes into domain-invariant Anchors and domain-sensitive Variants at the input level. scHelix utilizes a dual-stream sparse diffusion encoder equipped with stop-gradient graph caching to efficiently learn multi-scale structural representations. The core of our approach is a novel asymmetric Align-Refine-Fuse protocol: the unstable Variant stream is first aligned to the robust topology of the Anchor stream, followed by a conservative refinement phase where the Anchor stream absorbs denoised details via bounded residual gating. This divide-and-conquer architecture prevents shortcut learning and ensures robust batch removal without compromising the integrity of biological clusters. Extensive benchmarking demonstrates that scHelix outperforms state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2605_18576
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle scHelix: Asymmetric Dual-Stream Integration via Explicit Gene-Level Disentanglement
Yan, Xichen
Zang, Zelin
Chi, Changxi
Zhou, Jingbo
Yu, Chang
Wu, Jinlin
Cheng, Shenghui
Yang, Fuji
Luo, Jiebo
Lei, Zhen
Li, Stan Z.
Machine Learning
A critical challenge in single-cell RNA sequencing (scRNA-seq) integration is resolving the tension between eliminating batch effects and maintaining biological fidelity. While recent evidence indicates that batch effects manifest heterogeneously across genes, most existing methods process the transcriptome uniformly, frequently resulting in over-correction and loss of subtle biological signals. To address this, we present scHelix, a dataset-adaptive framework that fundamentally changes how features are processed by explicitly partitioning genes into domain-invariant Anchors and domain-sensitive Variants at the input level. scHelix utilizes a dual-stream sparse diffusion encoder equipped with stop-gradient graph caching to efficiently learn multi-scale structural representations. The core of our approach is a novel asymmetric Align-Refine-Fuse protocol: the unstable Variant stream is first aligned to the robust topology of the Anchor stream, followed by a conservative refinement phase where the Anchor stream absorbs denoised details via bounded residual gating. This divide-and-conquer architecture prevents shortcut learning and ensures robust batch removal without compromising the integrity of biological clusters. Extensive benchmarking demonstrates that scHelix outperforms state-of-the-art methods.
title scHelix: Asymmetric Dual-Stream Integration via Explicit Gene-Level Disentanglement
topic Machine Learning
url https://arxiv.org/abs/2605.18576