Genetic-to-Chemical Perturbation Transfer Learning Through Unified Multimodal Molecular Representations
Preprint
Abstract
Artificial Intelligence virtual cell (AIVC) holds transformative potential for biomedical research. Central to this vision is the systematic modeling of genetic and chemical perturbation phenotypes to accurately predict cellular dynamic states from diverse interventions. However, disparities in screening agents, library scales, experimental technologies, and data production efficiency hinder the integration, modeling, and analysis of the cross-data. Here we present UniPert-G2CP , a two-phase deep learning approach comprising i) UniPert, a multimodal molecular representation model that bridges genetic and chemical domains, and ii) G2CP ( Genetic-to-Chemical Perturbation transfer learning), which systematically transforms CRISPR screen-based genetic insights into chemical perturbation modeling for cost-effective in silico drug screening. UniPert not only encodes multimodal perturbagens into a unified functionally interpretable sematic embedding space, but also improves phenotypic effect prediction for previously unseen gene perturbations and drug treatments. Building upon UniPert, G2CP successfully modeled large-scale cellular post-perturbation states spanning 4,994 gene and 7,821 compound perturbagens, while reducing modeling data costs by over 60%. We demonstrate that UniPert-G2CP enables efficient, generalizable simulations of multicellular, multi-domain perturbation cause-effect spaces, revealing differential cellular biological causality and informing mechanism-driven therapy. UniPert-G2CP opens new avenues for biological causal foundation model building, AIVC creation, and AI-powered precision medicine.Keywords
This publication has 136 references indexed in Scilit:
- Toward a foundation model of causal cell and tissue biology with a Perturbation Cell and Tissue AtlasCell, 2024
- A mini-review on perturbation modelling across single-cell omic modalitiesComputational and Structural Biotechnology Journal, 2024
- Next-generation forward genetic screens: uniting high-throughput perturbations with single-cell analysisTrends in Genetics, 2023
- The proteomic landscape of genome-wide genetic perturbationsCell, 2023
- Comparison of three zinc binding groups for HDAC inhibitors – A potency, selectivity and enzymatic kinetics studyBioorganic & Medicinal Chemistry Letters, 2022
- A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 ProfilesCell, 2017
- Defining a Cancer Dependency MapCell, 2017
- Perturb-Seq: Dissecting Molecular Circuits with Scalable Single-Cell RNA Profiling of Pooled Genetic ScreensCell, 2016
- Prediction of protein cellular attributes using pseudo‐amino acid compositionProteins: Structure, Function, and Bioinformatics, 2001
- Identification of common molecular subsequencesJournal of Molecular Biology, 1981