Article

A confound-diagnostic toolkit for in silico perturbation with single-cell foundation models

2026-08-07

Abstract excerpt

<h4>Summary</h4> Deleting a gene token from a cell’s input sequence offers a convenient native strategy for in silico perturbation, but the resulting embedding delta may not represent a biological knockout response. Apparent effects can instead reflect gene identity, universal responsiveness, limited tokenization coverage, library-size contamination, or circular state scoring. Here, we present a confound-diagnost...

Topics

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

Identifiers and source

Literature Corpus work: 5dfa0da6-6a95-5224-b8ed-7744b60cd150
DOI: 10.64898/2026.08.04.732812

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

A confound-diagnostic toolkit for in silico perturbation with single-cell foundation modelsDOI 10.64898/2026.08.04.732812Read

Benchmarking AI Models for <i>In Silico</i> Gene Perturbation of CellsSemantic proximity Read

X-Atlas/Orion: Genome-wide Perturb-seq Datasets via a Scalable Fix-Cryopreserve Platform for Training Dose-Dependent Biological Foundation ModelsSemantic proximity Read

Cell Type-Agnostic Optical Perturbation Screening Using Nuclear In-Situ Sequencing (NIS-Seq)Semantic proximity Read

Three Classes of Confound in Gene-Regulatory-Network Inference: A Systematic Audit and Open-Source Diagnostic ToolkitSemantic proximity Read

Benchmarking ambient RNA removal across droplet and well-plate platforms reveals artificial count generation as a critical failure mode of scAR and CellClearSemantic proximity Read

Genome mutations and environmental perturbations reshape the epigenomic topology to drive functional collapse in digital cellsSemantic proximity Read

Stellar quality control for single-cell image-based profiling with coSMicQCSemantic proximity Read

Guide RNA library representation results in gene essentiality prediction bias in genome-wide CRISPR screensSemantic proximity Read

pertTF: context-aware AI modeling for genome-scale and cross-system perturbation predictionSemantic proximity Read

Virtual-cell verification enables self-auditing AI discovery for immune rejuvenationSemantic proximity Read

Method Choice, Not Biology, Determines In Silico Perturbation Results: A Systematic Evaluation of Eight Methods Across Four DatasetsSemantic proximity Read

The effect of background noise and its removal on the analysis of single-cell expression data.Semantic proximity Read