Accurately modeling RNase H-mediated antisense oligonucleotide efficacy
12 affiliations
- 1Big Data Institute and Centre for Human Genetics, University of Oxford, Oxford OX3 7LF, UK
- 2Department of Paediatrics, University of Oxford, Oxford OX3 7TY, UK
- 3Institute of Developmental and Regenerative Medicine, IMS-Tetsuya Nakamura Building, Old Road Campus, Oxford OX3 7TY, UK
- 4Department of Paediatrics, University of Oxford, OX3 7TY Oxford, United Kingdom
- 5MRC Nucleic Acid Therapy Accelerator, Harwell Research and Innovation Campus, Didcot OX11 0FA, UK
- 6MRC Nucleic Acid Therapy Accelerator, Harwell Research and Innovation Campus, OX11 0FA Didcot, United Kingdom
- 7Broad Center for Mendelian Genomics, Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
- 8Big Data Institute and Centre for Human Genetics, University of Oxford, OX 7LF Oxford, United Kingdom
- 9Oxford Harrington Rare Disease Centre, IDRM, Old Road Campus, Oxford OX3 7TY, UK
- 10Department of Paediatrics, University of Oxford, OX3 7TY Oxford, United Kingdom 2MRC Nucleic Acid Therapy Accelerator, Harwell Research and Innovation Campus, OX11 0FA
- 11Department of Psychiatry and Behavioral Sciences, UCSF Weill Institute for Neurosciences, University of California, San Francisco, San Francisco, CA 94178, USA
- 12New York Genome Center, New York, NY 10013, USA
Abstract
Antisense oligonucleotides (ASOs) are a powerful class of drugs with the potential to treat a wide range of human diseases. However, the prediction of ASO efficacy remains challenging, as large-scale and costly experimental screens are typically required to identify optimal candidates for a specific therapeutic target. To address this challenge, we compiled ASO Atlas, a database comprising 188,521 RNase H-mediated ASO sequences targeting 334 unique genes with corresponding knockdown efficacy measurements extracted from published patents. Using ASO Atlas, we trained OligoAI, a deep learning model capable of jointly modelling RNA target context, ASO sequence, sugar and backbone chemistries, and dosage to predict in vitro efficacy. We experimentally validated OligoAI by targeting KCNT2, achieving a 5.72-fold reduction in screening effort compared to random selection. ASO Atlas provides the first systematic resource to rigorously evaluate hypotheses regarding key parameters in ASO design, including sequence composition, chemical modifications, and target region selection. Both ASO Atlas and OligoAI have been made freely accessible through an online web-tool with the aim of facilitating the accelerated optimisation of ASO design.
Related
- Biophysics-informed computational framework for predicting antisense oligonucleotide knockdown efficacy under small-data constraints.
- Cell-resolved pharmacodynamic profiling reveals tissue- and cell-population-specific activity of antisense oligonucleotide chemical conjugates in vivo
- Fundamentals on the Kinetic and Thermodynamic Analysis of Oligonucleotide DNA Hybridization by Surface Plasmon Resonance: A Guide for HIF1α Antisense Design.
- Enhancing the efficacy of siRNA Antibody Oligonucleotide Conjugates (AOCs) through chemical design
- Inhibition of 5’-tiRNAGly restores anti-PD-1 immunotherapy efficacy via DLST-mediated OGDHL succinylation in gastric cancer
BibTeX
@article{hill2026accurately,
title = {{Accurately modeling RNase H-mediated antisense oligonucleotide efficacy}},
author = {Barney Hill and Maisie R. Jaques and R. Nair and Nicola Whiffin and Matthew J. A. Wood and Stephan Sanders and Peter L. Oliver and Alyssa C. Hill and Carlo Rinaldi},
journal = {Molecular Therapy — Nucleic Acids},
year = {2026},
month = jul,
doi = {10.1016/j.omtn.2026.103004},
url = {https://doi.org/10.1016/j.omtn.2026.103004},
note = {CC BY},
}
RIS
TY - JOUR
TI - Accurately modeling RNase H-mediated antisense oligonucleotide efficacy
AU - Barney Hill
AU - Maisie R. Jaques
AU - R. Nair
AU - Nicola Whiffin
AU - Matthew J. A. Wood
AU - Stephan Sanders
AU - Peter L. Oliver
AU - Alyssa C. Hill
AU - Carlo Rinaldi
PY - 2026
DA - 2026/07/07
JO - Molecular Therapy — Nucleic Acids
DO - 10.1016/j.omtn.2026.103004
UR - https://doi.org/10.1016/j.omtn.2026.103004
AB - Antisense oligonucleotides (ASOs) are a powerful class of drugs with the potential to treat a wide range of human diseases. However, the prediction of ASO efficacy remains challenging, as large-scale and costly experimental screens are typically required to identify optimal candidates for a specific therapeutic target. To address this challenge, we compiled ASO Atlas, a database comprising 188,521 RNase H-mediated ASO sequences targeting 334 unique genes with corresponding knockdown efficacy measurements extracted from published patents. Using ASO Atlas, we trained OligoAI, a deep learning model capable of jointly modelling RNA target context, ASO sequence, sugar and backbone chemistries, and dosage to predict in vitro efficacy. We experimentally validated OligoAI by targeting KCNT2, achieving a 5.72-fold reduction in screening effort compared to random selection. ASO Atlas provides the first systematic resource to rigorously evaluate hypotheses regarding key parameters in ASO design, including sequence composition, chemical modifications, and target region selection. Both ASO Atlas and OligoAI have been made freely accessible through an online web-tool with the aim of facilitating the accelerated optimisation of ASO design.
SN - 2162-2531
ER -