AblateCell: A Reproduce-then-Ablate Agent for Virtual Cell Repositories

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xia, Xue, Yao, Chengkai, Tsoi, Mingyu, Mao, Xinjie, Huang, Wenxuan, Wei, Jiaqi, Wu, Hao, Tan, Cheng, Yu, Lang, Yang, Yuejin, Liu, Mengdi, Sun, Siqi, Gao, Zhangyang
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915970041249792
author Xia, Xue
Yao, Chengkai
Tsoi, Mingyu
Mao, Xinjie
Huang, Wenxuan
Wei, Jiaqi
Wu, Hao
Tan, Cheng
Yu, Lang
Yang, Yuejin
Liu, Mengdi
Sun, Siqi
Gao, Zhangyang
author_facet Xia, Xue
Yao, Chengkai
Tsoi, Mingyu
Mao, Xinjie
Huang, Wenxuan
Wei, Jiaqi
Wu, Hao
Tan, Cheng
Yu, Lang
Yang, Yuejin
Liu, Mengdi
Sun, Siqi
Gao, Zhangyang
contents Systematic ablations are essential to attribute performance gains in AI Virtual Cells, yet they are rarely performed because biological repositories are under-standardized and tightly coupled to domain-specific data and formats. While recent coding agents can translate ideas into implementations, they typically stop at producing code and lack a verifier that can reproduce strong baselines and rigorously test which components truly matter. We introduce AblateCell, a reproduce-then-ablate agent for virtual cell repositories that closes this verification gap. AblateCell first reproduces reported baselines end-to-end by auto-configuring environments, resolving dependency and data issues, and rerunning official evaluations while emitting verifiable artifacts. It then conducts closed-loop ablation by generating a graph of isolated repository mutations and adaptively selecting experiments under a reward that trades off performance impact and execution cost. Evaluated on three single-cell perturbation prediction repositories (CPA, GEARS, BioLORD), AblateCell achieves 88.9% (+29.9% to human expert) end-to-end workflow success and 93.3% (+53.3% to heuristic) accuracy in recovering ground-truth critical components. These results enable scalable, repository-grounded verification and attribution directly on biological codebases.
format Preprint
id arxiv_https___arxiv_org_abs_2604_19606
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle AblateCell: A Reproduce-then-Ablate Agent for Virtual Cell Repositories
Xia, Xue
Yao, Chengkai
Tsoi, Mingyu
Mao, Xinjie
Huang, Wenxuan
Wei, Jiaqi
Wu, Hao
Tan, Cheng
Yu, Lang
Yang, Yuejin
Liu, Mengdi
Sun, Siqi
Gao, Zhangyang
Artificial Intelligence
Multiagent Systems
Systematic ablations are essential to attribute performance gains in AI Virtual Cells, yet they are rarely performed because biological repositories are under-standardized and tightly coupled to domain-specific data and formats. While recent coding agents can translate ideas into implementations, they typically stop at producing code and lack a verifier that can reproduce strong baselines and rigorously test which components truly matter. We introduce AblateCell, a reproduce-then-ablate agent for virtual cell repositories that closes this verification gap. AblateCell first reproduces reported baselines end-to-end by auto-configuring environments, resolving dependency and data issues, and rerunning official evaluations while emitting verifiable artifacts. It then conducts closed-loop ablation by generating a graph of isolated repository mutations and adaptively selecting experiments under a reward that trades off performance impact and execution cost. Evaluated on three single-cell perturbation prediction repositories (CPA, GEARS, BioLORD), AblateCell achieves 88.9% (+29.9% to human expert) end-to-end workflow success and 93.3% (+53.3% to heuristic) accuracy in recovering ground-truth critical components. These results enable scalable, repository-grounded verification and attribution directly on biological codebases.
title AblateCell: A Reproduce-then-Ablate Agent for Virtual Cell Repositories
topic Artificial Intelligence
Multiagent Systems
url https://arxiv.org/abs/2604.19606