Multiplexed In Vivo Base Editing Identifies Functional Gene-Variant-Context Interactions

Abstract

A bioRxiv preprint (Feb 2025, with PMC publication 2026) demonstrates multiplexed in vivo base editing as a tool for systematically identifying functional gene-variant-context interactions. The work pushes base editing from a single-target therapeutic modality toward a high-throughput functional genomics platform — enabling in vivo screening of variant effects across genes and contexts.

Key Contributions

• Multiplexed in vivo base editing — multiple variants edited per animal, enabling systematic screens.
• Gene-variant-context interaction mapping — identifies how the same variant has different functional effects depending on genetic context.
• High-throughput functional genomics in vivo — bridges the gap between in vitro variant screens (CRISPR-Cas9 KO screens) and the more constrained in vivo setting.
• Methodology platform — usable for many disease and tissue contexts.

Methodology

• Multiple gRNAs / variants delivered simultaneously to model animals.
• Base editor mRNA + lipid-nanoparticle (LNP) delivery for hepatocyte targeting.
• Sequencing-based readout of editing outcomes per cell, per variant.
• Statistical analysis of variant-effect distributions across genetic contexts.

Results

• Multiplexed editing achieved at scales meaningfully larger than single-variant approaches.
• Identified gene-variant-context interactions that would be missed in single-target studies.
• Validates LNP-delivered base editor mRNA as a tractable in vivo platform.

Limitations

• Variant-context interactions are complex; statistical power per variant decreases as multiplexing increases.
• LNP delivery primarily targets hepatocytes — extension to other tissues requires alternative formulations.
• Translation from animal-model variant effects to human disease requires careful mapping.

Full Content

The strategic framing of this work is that gene editing infrastructure (delivery, base editing, prime editing, multiplexed gRNAs) is becoming a general-purpose biology platform, not just a therapeutic-development pathway. As the platform matures, it serves two parallel functions:

1. Therapeutics (Casgevy, Verve, Beam, RBM20 cardiac prime editing): single-target precision interventions for known disease variants.
2. Functional genomics in vivo: systematic mapping of variant effects, gene-context interactions, and disease mechanisms.

These functions reinforce each other — better functional understanding identifies which targets to pursue therapeutically, while therapeutic-grade delivery technology enables more sophisticated functional studies.

For genomics research, the 2026 frontier is:

• In vivo CRISPR screens at scale (this paper) — push variant effect mapping from cell lines to whole organism.
• Tissue-specific delivery (RBM20 cardiac prime editing) — eliminate confounding off-target tissue effects.
• AI-designed gene editors (OpenCRISPR-1) — expand the editor toolbox.
• DNA methylation-based gene control (UNSW CRISPR without cutting) — silence/activate genes without permanent edits.

The convergence of these threads is a unified gene-editing platform: AI designs the editor, multiplexed delivery hits the target tissue, the editor performs the precise change without off-target effects, and the result is screened functionally before therapeutic application.

Source: bioRxiv 2025.02.23.639770 — Multiplexed in vivo base editing identifies functional gene-variant-context interactions, with PMC publication 2026