aiqbench analysis
This research advances biological machine learning by overcoming synthesis bottlenecks through continuous-space variational models, enabling the physical manufacturing of diverse DNA sequences like enzymes and antibodies. By applying post-training quantization to satisfy wetware constraints, the method achieves a superior quality-diversity trade-off that translates effectively from simulation to in vitro performance.
Continuous Variational Synthesis
arXiv:2609.35083v1 Announce Type: cross Abstract: Biological machine learning was long bottlenecked by the ability to synthesize designed DNA. Variational synthesis models control chemical reactions to physically manufacture quadrillions of designed sequences in DNA. However, training these generative models is challenging: constraints on chemical synthesis can force many parameters into a discrete space, limiting the ability to pre-train and fine-tune. In this article we train ``free'' variational synthesis models using stochastic gradient descent in continuous space, and then discretize with post-training quantization to impose hardware and wetware constraints. This enables variational synthesis models to satisfy stringent reward criteria, while still synthesizing diverse designs, achieving a strictly dominating quality-diversity Pareto frontier. We demonstrate by training variational synthesis models of enzymes, peptides, antibody CDRH3s, and regulatory DNA elements. In silico performance is maintained in vitro.
Abstract:Biological machine learning was long bottlenecked by the ability to synthesize designed DNA. Variational synthesis models control chemical reactions to physically manufacture quadrillions of designed sequences in DNA. However, training these generative models is challenging: constraints on chemical synthesis can force many parameters into a discrete space, limiting the ability to pre-train and fine-tune. In this article we train ``free'' variational synthesis models using stochastic gradient descent in continuous space, and then discretize with post-training quantization to impose hardware and wetware constraints. This enables variational synthesis models to satisfy stringent reward criteria, while still synthesizing diverse designs, achieving a strictly dominating quality-diversity Pareto frontier. We demonstrate by training variational synthesis models of enzymes, peptides, antibody CDRH3s, and regulatory DNA elements. In silico performance is maintained in vitro.
Biomolecules (q-bio.BM); Machine Learning (cs.LG)
arXiv:2609.35083 [q-bio.BM]
(or arXiv:2609.35083v1 [q-bio.BM] for this version)
https://doi.org/10.48550/arXiv.2609.35083 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 223 words
Related coverage
ImprovedVBGS: Real-time Continual Variational Bayes Gaussian Splatting
arXiv · Sep 2, 2026
News
Quantum Computing Report · Jul 25, 2026
Foundations of Diffusion Models in General State Spaces: A Self-Contained Introduction
arXiv · Aug 20, 2026
FACET: Factorized Asymmetric Conditioning for Efficient Transport in High-Fidelity Fluorescence Microscopy Synthesis
arXiv · Oct 6, 2026