Deaminet 2026, held in Palm Springs in late January, brought together researchers studying deaminase enzymes across disciplines: from structural biologists resolving APOBEC proteins at atomic resolution, to cancer biologists dissecting mutational processes, to medical researchers advancing base and prime editing toward the clinic. Across talks spanning mechanism, off-target biology, and in vivo delivery, one theme stood out clearly: as our understanding of how these editors work deepens, the tools themselves are evolving and being adopted faster than ever.
The field in numbers
At Deaminet, we presented an analysis of 14 years of CRISPR plasmid deposition and distribution data at Addgene. Since 2012, Addgene has distributed tens of thousands of CRISPR tools, with base and prime editors representing two of the fastest growing segments of the collection. Base editor deposits peaked in 2020, followed by prime editors in 2021, yet both collections have continued to expand steadily.
One striking pattern is the longevity of community demand: CRISPR plasmids deposited as early as 2013 still account for ~4% of all 2025 distributions. At the same time, several editors deposited in 2024 already rank among the most requested base editing tools of 2025. Together, these trends reflect both the durability of foundational tools and the community’s ability to rapidly identify and adopt new innovations.
The talks at Deaminet help explain why certain tools rise to the top, and what the next wave may look like. Here are six take-aways from the meeting.
1. Mechanistic deep dives: Understanding deaminases at atomic resolution
PE6d: Processivity gains and overextension trade-offs
The prime editor PE6d exhibits substantially increased processivity and editing efficiency relative to PE2 (Doman et al., 2023), but this improvement comes with a trade-off: overextension past the reverse transcriptase (RT) template, resulting in unintended insertions. Audrone Lapinaite (UC Irvine) presented work dissecting the underlying molecular mechanism of this pattern using a combination of biochemical, structural assays, cryo-EM, and computational approaches.
PE6d’s higher editing efficiency is not explained by dNTP availability alone, indicating an intrinsic shift in enzyme processivity. Structural data implicate specific residues in the reverse transcriptase polymerase domain as likely determinants of the enzyme’s enhanced processivity. Ongoing studies are evaluating how these residues influence both efficiency and fidelity.
 |
|
Figure 1: The first steps of prime editing. DNA release and heteroduplex resolution are not shown. Created with BioRender.com. |
PE6d, deposited by David Liu’s lab, is among the 2025 most-distributed prime editors.
ABE8e: How TadA8e achieves its remarkable efficiency
The adenine base editor ABE8e (Richter et al., 2020) remains among the most-requested base editors at Addgene in 2025. Giulia Palermo (UC Riverside) presented molecular simulation studies revealing that TadA8e dimerization, together with a unique Cas9-DNA “locking” interaction, are critical for its high deamination efficiency (Arantes et al., 2024). By comparing conformational selection and induced fit binding models, her team identified conformational selection as the dominant pathway for ABE8e activity. These mechanistic insights are now being incorporated into diffusion models to guide rational design of a new generation of deaminases.
|
|
|
Figure 2: Structure of ABE8e*A showing dimerized TadA subunits, with the catalytic domain in dark red and the docking domain in pink. Reproduced from Arantes et al. (2024) under a CC BY-NC 4.0 License. |
2. Precision and safety: A new generation of off-target detection
As base and prime editors diversify, off-target editing is becoming more diverse, mechanistically complex, and harder to predict. Several talks at Deaminet highlighted that standard guide‑directed off‑target workflows often fail to capture newly described editing outcomes or meet emerging clinical requirements. Together, these presentations underscored the need for more sensitive, modality‑matched approaches to off‑target characterization.
beCasKAS: Linking Cas9 binding to editing outcomes
Tong Wang (Stanford) presented beCasKAS (Wang et al., 2025), an in cellulo assay that simultaneously detects Cas9-mediated DNA unwinding (R loops) and resulting deaminase edits. Applied in primary T cells, the method nominates over 460-fold more off target sites than edit-only approaches and revealed distinct noncoding off-target landscapes for ABE8e and PAM-relaxed ABE8e-SpRY. Using deep‑learning models, the team ranked noncoding off‑target edits by their predicted effects on chromatin and gene regulation. These results highlight that Cas9 binding and editing outcomes can diverge substantially, reinforcing the need for assays that capture multiple layers of editor behavior.
 |
|
Figure 3: Labeling of Cas9-mediated R loops. Reproduced from Wang et al. (2025) under a CC BY 4.0 License. |
Inrich-seq: Capturing the editing intermediate directly
While ABEs can theoretically correct nearly half of all pathogenic SNVs, faithful detection of ABE off-targets remains challenging. Siyuan Zou (U. Chicago) presented Inrich-seq, a method that captures the ABE intermediate inosine directly in cellulo using a novel chemistry. This approach achieves ultra-low false-positive (0–9%) and false-negative (0–19%) rates. In clinically relevant HSPC models, Inrich-seq revealed numerous high-frequency off-target sites in the HBG promoter that were missed by standard workflows, highlighting the added value of directly capturing editing intermediates.
Multiplexing approaches to editing outcomes profiling
Two new methods to assess editing profiling of multiple guides in one experiment were presented. BESTIE (Nicola Osgood, UC San Diego) enables pooled screening of hundreds of gRNA–base editor combinations in a single experiment. ProBE (Emily King, Kleinstiver Lab, Harvard) is a 96-well platform for parallel characterization of hundreds of base editor proteins across kinetics, editing windows, sequence preferences, and edit types.
3. From bench to bedside: Therapeutic translation is accelerating
Base editors and prime editors are steadily advancing through clinical evaluation. At Deaminet 2026, several talks presented in vivo efficacy data, comprehensive off-target assessments, and IND-enabling studies.
PERT: A disease-agnostic prime editing strategy
Steven Erwood (David Liu’s lab, Broad Institute), presented PERT: prime editing readthrough of premature termination codons (Pierce et al., 2025). The approach uses prime editing to permanently convert an endogenous tRNA into a suppressor tRNA, enabling readthrough of nonsense mutations without lifelong dosing. Approximately 11% of all pathogenic alleles are caused by premature stop codons.
By screening 18,000 pegRNAs across all 418 human tRNAs, they identified the strongest tRNA candidates. Engineered Leu suppressor tRNAs achieved ~30% protein rescue in cell models of Batten and Tay-Sachs disease. In vivo, AAV-delivered PERT components extensively rescued disease pathology in a Hurler syndrome mouse model.
Conceptually, PERT reframes nonsense mutations from a gene-specific problem to a stop codon–class problem, offering a potentially scalable therapeutic strategy.
 |
|
Figure 4: Schematic of PERT. Reproduced from Pierce et al. (2025) under a CC BY 4.0 License. |
oDNA: Immune-evasive DNA donors for pan-mutation editing
Benjamin Kleinstiver (Massachusetts General Hospital / Harvard Medical School) presented oDNA: circular single-stranded DNA with short double-stranded regions just long enough for recombinase binding, but too short to activate cGAS-sensing immune response.
In vivo conventional dsDNA caused toxicity at low doses, whereas oDNA was well tolerated at doses fourfold higher and achieved ~1% non-viral liver integration. The oDNA approach addresses a key barrier for pan-mutation editing strategies by enabling large sequence integration without triggering immunotoxicity.
PM577a: Prime editing for Wilson's disease
Maria Collier (Prime Medicine) presented one of the most clinically advanced programs at the meeting: PM577a, a lipid nanoparticle-formulated prime editor delivered to the liver to correct the ATP7B p.H1069Q mutation in Wilson’s disease. Results showed >80% precise correction and restoration of copper homeostasis to wild-type levels.
Across a comprehensive suite of complementary off-target assessment assays, including PEG-seq (Stewart-Ornstein et al., 2025), DNA-FISH for chromosomal structural variant analysis, targeted sequencing of computationally nominated sites, and validation in patient-derived iPSCs, PM577a showed no detectable off-target editing or chromosomal abnormalities. Together, these data illustrate how rigorously editor safety can now be assessed as programs move toward the clinic. Prime Medicine announced that IND/CTA filing for PM577a is on track for the first half of 2026.
4. RNA editing comes into its own
RNA editing has moved decisively beyond proof of concept, with multiple platforms now demonstrating durability, tunability, and clinical relevance.
AIMers
Deepak Shivalila described Wave Life Science’s GalNAc-conjugated AIMer platform, which engages endogenous ADAR for hepatocyte-specific A-to-I RNA editing. The lead program, WVE-006, corrects the SERPINA1 Z mutation in alpha-1 antitrypsin deficiency (AATD) and is currently in Phase 1b/2a trials, with published clinical data showing durable production of functional AAT protein.
Adaptamers
Daniel Bryant (Prashant Mali's lab, UC San Diego) presented adaptamers, compact (<120 bp) RNA switches that couple small-molecule responsive aptamers with endogenous ADAR recruitment to regulate transgene expression (Bryant et al., 2025). In mice, AAV-delivered adaptamer-controlled FGF21 expression reversed obesity and metabolic comorbidities, demonstrating tunable, re-dosable control of therapeutic transgenes via endogenous ADAR in vivo.
Editopes
Nina Papavasiliou described using short guide RNAs to recruit endogenous ADAR1 to mRNA targets, generating edited epitopes: “editopes”, which are recognized by the immune system as tumor-specific neoantigens (Pecori et al., 2025). Even low editing efficiencies were sufficient to prime robust T cell responses, converting immunologically cold tumors into immunotherapy targets.
5. Deaminases in cancer: Dual roles as mutagens and therapeutic targets
Many of the deaminases that power modern base editors, particularly APOBEC family members, also function endogenously, where their activity shapes genome evolution in both health and disease. Several talks at Deaminet highlighted how dysregulated APOBEC activity drives cancer progression, offering important context for understanding the risks, constraints, and potential therapeutic leverage points of these enzymes.
Pedro Ortega (UC Irvine) showed that APOBEC3B activity at replication forks induces APE1-dependent abasic site cleavage, and PARP1 hyperactivation, forming nuclear condensates that disrupt RNA processing and translation (Ortega et al., 2025). These findings position APOBEC3B as an active driver able to reshape tumor cell biology and point to high APOBEC3B tumors as potential targets for synthetic lethal therapeutic strategies.
Sarat Chandarlapaty (MSKCC) reported that APOBEC3-dominant breast tumors resist endocrine therapies and CDK4/6 inhibitors, consistent with recent large-cohort studies linking APOBEC signatures to shorter progression-free survival (Gupta et al., 2025). Together, these data support APOBEC activity as both a prognostic marker and a potential stratification marker in treatment decisions.
Shelby Devenport (Washington University) demonstrated that even subclonal A3A expression (~10% of cells) can reshape entire ovarian tumors via IL 6–mediated paracrine signaling, explaining why tumors with low overall APOBEC burden can still exhibit aggressive behavior.
Finally, Alex Wallen (UPenn) showed APOBEC3B-driven mutagenesis increases dendritic cell infiltration in head and neck cancer models, converting immune-cold tumors to immune-responsive ones and potentially enabling new immunotherapy combinations.
Taken together, these studies highlight the dual nature of deaminases as both liabilities and opportunities, an important reminder that the same chemistry enabling precise genome editing can profoundly reshape cellular biology when misregulated.
6. Where the field is going
The speed of clinical translation in genome and RNA editing is remarkable. Tools deposited with Addgene in 2024 are already amongst the most distributed of 2025, and many of the programs presented at Deaminet 2026 are advancing through IND-enabling studies using editing tools developed only a few years ago. The gap between first publication and widespread adoption has never been shorter.
As cryo-EM structures, molecular simulations, and structure informed diffusion models become more widely available, enzyme engineering is accelerating. In parallel, newly characterized editing outcomes are driving the need for increasingly precise, modality appropriate off-target assessment methods.
RNA and DNA editing are increasingly seen as complementary approaches, ranging from transient RNA correction with AIMers to permanent genomic changes enabled by prime editing.
As the toolkit expands to include novel deaminase families (mSCD, DYW), new Cas9 orthologs (ePsCas9, NmeCas9), CRISPR-free approaches (MRGE, oDNA), and RNA editing platforms, open sharing of these tools remains essential.
Every plasmid deposited and every protocol shared enables scientists worldwide to build on existing tools, accelerating progress across the gene editing field, from fundamental mechanistic insights to translational applications.
If you presented at Deaminet 2026 and have not yet deposited your plasmids, deposit with Addgene to enable rapid, worldwide open access and maximize your research’s impact!
This blog post was written by Addgenie Elena Minones-Moyano, PhD.
References and Resources
References
Arantes, P. R., Chen, X., Sinha, S., Saha A., Patel A.C., Sample M., Nierzwicki L., Lapinaite A., Palermo G. (2024). Dimerization of the deaminase domain and locking interactions with Cas9 boost base editing efficiency in ABE8e. Nucleic Acids Research, 52(22):13931-13944. https://doi.org/10.1093/nar/gkae1066.
Bryant, J., Herron, L., Doctor, Y., McRae, F., Wang, J., Hirsch, T., Portell, A., Kumar, S., & Mali, P. (2025). Regulatable in vivo gene expression via adaptamers. bioRxiv. https://doi.org/10.64898/2025.12.21.695777
Doman, J. L., Pandey, S., Neugebauer, M. E., An, M., Davis, J. R., Randolph, P. B., McElroy, A., Gao, X. D., Raguram, A., Richter, M. F., Everette, K. A., Banskota, S., Tian, K., Tao, Y. A., Tolar, J., Osborn, M. J., & Liu, D. R. (2023). Phage-assisted evolution and protein engineering yield compact, efficient prime editors. Cell, 186(18), 3983–4002. https://doi.org/10.1016/j.cell.2023.07.039
Gupta, A., Gazzo, A., Selenica, P., Safonov, A., Pareja, F., da Silva, E. M., Selenica, P., & Chandarlapaty, S. (2025). APOBEC3 mutagenesis drives therapy resistance in breast cancer. Nature Genetics, 57(6), 1452–1462. https://doi.org/10.1038/s41588-025-02187-1
Ortega, P., Bournique, E., Li, J., Sanchez, A., Santiago, G., Harris, B. R., Striepen, J., Maciejowski, J., Green, A. M., & Buisson, R. (2025). Mechanism of DNA replication fork breakage and PARP1 hyperactivation during replication catastrophe. Science Advances, 11(16), eadu0437. https://doi.org/10.1126/sciadv.adu0437
Pecori, R., Casati, B., Merdler-Rabinowicz, R., Landesman, N., Sanghvi, K., Zens, S., Kipfstuhl, K., Pinamonti, V., Arnold, A., Lindner, J. M., Platten, M., Offringa, R., Carretero, R., Ruppin, E., Levanon, E. Y., & Papavasiliou, F. N. (2025). Employing RNA editing to engineer personalized tumor-specific neoantigens (editopes). bioRxiv. https://doi.org/10.1101/2023.03.16.532918
Pierce, S. E., Erwood, S., Oye, K., An, M., Krasnow, N., Zhang, E., Raguram, A., Seelig, D., Osborn, M. J., & Liu, D. R. (2025). Prime editing-installed suppressor tRNAs for disease-agnostic genome editing. Nature, 648(8092), 191–202. https://doi.org/10.1038/s41586-025-09732-2
Richter, M. F., Zhao, K. T., Eton, E., Lapinaite, A., Newby, G. A., Thuronyi, B. W., Wilson, C., Koblan, L. W., Zeng, J., Bauer, D. E., Doudna, J. A., & Liu, D. R. (2020). Phage-assisted evolution of an adenine base editor with improved Cas domain compatibility and activity. Nature Biotechnology, 38(7), 883–891. https://doi.org/10.1038/s41587-020-0453-z
Stewart-Ornstein, J., Irby, M. J., Lilieholm, M. K., Laparte, D., Collier, M. D., Aunins, T., Harjanto, D., Chang, A. N., Reyon, D., & Duffield, J. S. (2025). 3′-end ligation sequencing is a sensitive method to detect DNA nicks at potential sites of off-target activity induced by prime editors. Genome Research, 35(9), 2064–2075. https://doi.org/10.1101/gr.280164.124
Wang, T., Jessa, S., Marinov, G. K., Klemm, S., Kundaje, A., & Greenleaf, W. J. (2025). Sensitive, direct detection of non-coding off-target base editor unwinding and editing in primary cells. bioRxiv. https://doi.org/10.1101/2025.09.25.678665
Additional resources on the Addgene blog
- Prime Editing: Adding Precision and Flexibility to CRISPR Editing
- The Advances Behind The World's First Personalized CRISPR Treatment
Resources on addgene.org
- CRISPR Plasmids: Addgene's collection of CRISPR plasmids
- CRISPR Guide: Essential background information on CRISPR and the basics for planning your first CRISPR experiment
- CRISPR 101 eBook: A comprehensive CRISPR resource based on our blog series
Topics: CRISPR, Cancer, Base Editing, CRISPR Therapeutic Applications, Other CRISPR Tools
Leave a Comment