Related Experiment Video
Updated: Jun 10, 2025
08:09
Mapping the Binding Site of an Aptamer on ATP Using MicroScale Thermophoresis
Published on: January 7, 2017
10.6K
AptaDiff: de novo design and optimization of aptamers based on diffusion models
Zhen Wang1,2, Ziqi Liu1,3, Wei Zhang1
1Hangzhou Institute of Medicine, Chinese Academy of Sciences, Hangzhou, 310018 Zhejiang, China.
Briefings in Bioinformatics
|October 21, 2024
Summary
AptaDiff, a novel computational method, designs and optimizes aptamers using diffusion models, overcoming limitations of traditional in vitro selection. This accelerates the discovery of high-affinity aptamers for molecular targets.
Area of Science:
- Computational Biology
- Biotechnology
- Bioinformatics
Background:
- Aptamers are nucleic acid ligands with high affinity and specificity for target molecules.
- Traditional aptamer discovery methods like SELEX are limited by library size and sequencing capabilities.
- Existing methods capture only a fraction of the theoretical sequence space, constraining aptamer discovery.
Purpose of the Study:
- To introduce AptaDiff, the first in silico aptamer design and optimization method based on diffusion models.
- To generate aptamers that surpass the limitations of high-throughput sequencing data.
- To optimize aptamer affinity through a guided generation process.
Main Methods:
- AptaDiff utilizes diffusion models for in silico aptamer design and optimization.
- Employs motif-dependent latent embeddings from variational autoencoders.
- Incorporates affinity-guided aptamer generation via Bayesian optimization.
Main Results:
- AptaDiff demonstrated superiority over existing methods in aptamer quality and fidelity across four datasets.
- Validated aptamers showed significantly increased binding affinity (87.9% and 60.2% RU boost) and decreased dissociation constants (3.6-fold and 2.4-fold KD decrease).
- Optimized aptamers exhibited enhanced binding affinity compared to SELEX-derived candidates.
Conclusions:
- AptaDiff effectively designs and optimizes aptamers beyond traditional experimental constraints.
- The method accelerates the discovery of high-affinity aptamers.
- AptaDiff represents a significant advancement in computational aptamer discovery.
