Learning Perturbation Effects Through Contrastive Alignment of Transcriptomics and Textual Embeddings

Tianyu Liu, Artur Szałata, Wenxin Long, Fabian J Theis, Lingzhou Xue, Hongyu Zhao

Track: Main track

Keywords: perturbation effects, foundation models, large language models, contrastive learning, multimodal

TL;DR: A method for learning perturbation effects

Abstract: Single-cell perturbation screens offer a scalable approach for characterizing the effects of genetic and chemical interventions on cellular state. However, most existing representation-learning methods are tailored to a single perturbation modality and fail to explicitly incorporate external semantic knowledge, which limits their ability to generalize across datasets and perturbation types. Here, we introduce PertOmni, a CLIP-style multimodal representation-learning framework that aligns transcriptomic perturbation signatures with text-derived embeddings of curated gene and compound descriptions. PertOmni jointly trains a shared transcriptomic encoder and dataset-specific text encoders using a masked contrastive objective that emphasizes within–cell-type discrimination while mitigating confounding effects arising from cell-type heterogeneity. We evaluate the produced joint embedding space on bi-directional retrieval, drug–gene interaction inference, and perturbation prediction across both small-molecule and CRISPRi perturbation datasets, and demonstrate consistent improvements over strong baseline methods.

AI Policy Confirmation: I confirm that this submission clearly discloses the role of AI systems and human contributors and complies with the ICLR 2026 Policies on Large Language Model Usage and the ICLR Code of Ethics.

Submission Number: 8