An accurate nucleic acid-small molecule docking framework via geometric deep learning with large-scale pretraining
arXiv:2606.05198v1 Announce Type: cross Abstract: Nucleic acids are increasingly recognized as therapeutic targets beyond conventional protein-centered drug discovery, yet accurate and efficient docking of small molecules to nucleic acid structures remains challenging. Physics-based docking methods often show limited accuracy and efficiency, whereas deep learning approaches are constrained by the scarcity of experimentally resolved nucleic acid-ligand complexes. Here, we present NucleoDock, a deep learning framework for nucleic acid-small molecule docking. To address data scarcity, NucleoDock
The scarcity of experimental data for nucleic acid-ligand complexes has historically limited the efficacy of deep learning in drug discovery, but advances in geometric deep learning and large-scale pretraining are now overcoming these limitations.
This development could significantly accelerate drug discovery by enabling more accurate and efficient identification of small molecules that target nucleic acids, expanding the scope of therapeutic targets beyond proteins.
The ability to accurately dock small molecules to nucleic acids via advanced AI fundamentally changes the approach to designing drugs for previously intractable targets, offering new avenues for therapeutic development.
- · Pharmaceutical companies
- · Biotechnology sector
- · AI compute providers
- · Patients with currently untreatable diseases
- · Traditional physics-based docking software companies
New drug candidates targeting nucleic acids are identified more rapidly and efficiently.
The cost and timeline for early-stage drug discovery decrease, leading to an increase in new therapeutic pipelines.
A shift in drug discovery priorities towards nucleic acid-based targets, potentially unlocking treatments for genetic disorders and viral infections on an unprecedented scale.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG