Risky SELEX discovery
The Systematic Evolution of Ligands by Exponential Enrichment (SELEX) process is traditionally labor-intensive and time-consuming, often with uncertain outcomes.
AptaBLE (Aptamer Binding LanguagE) is a transformer-based AI model that designs and optimizes aptamers in days, not months.
Aptamers dock with their targets through shape-specific contacts — weeks of wet-lab selection compressed into days of in-silico design.
5–15 kDa vs. 150–170 kDa antibodies.
Reduced immune response in vivo.
Predictable, controlled chemistry.
Chemical synthesis in vitro.
AI-designed binding profiles.
Across temperature and pH ranges.
Aptamers offer significant advantages as targeting ligands; however, their development for specific biological applications presents several challenges:
The Systematic Evolution of Ligands by Exponential Enrichment (SELEX) process is traditionally labor-intensive and time-consuming, often with uncertain outcomes.
Aptamers, particularly RNA aptamers, are prone to rapid degradation by nucleases, leading to short in vivo half-lives and limited therapeutic windows.
Aptamers selected in isolation may inadvertently bind to structurally similar compounds or epitopes, resulting in off-target effects or failure to bind endogenous targets effectively.
Aptamers offer immense potential as highly specific targeting ligands, but traditional wet-lab discovery methods are slow, expensive, and labor-intensive. To overcome these limitations, we developed AptaBLE (Aptamer Binding LanguagE), a cutting-edge, transformer-based AI model trained on proprietary data from protein-aptamer interactions. AptaBLE decodes these complex interactions, enabling us to design and optimize aptamers with unprecedented speed, accuracy, and efficiency.
Here's how the AptaBLE platform solves key challenges in traditional aptamer development:
AptaBLE revolutionizes this with AI-driven capabilities to 1) rapidly generate diverse pools of aptamer candidates, and 2) optimize existing aptamers to improve performance. This results in aptamer development that takes days instead of months, at a fraction of the cost.
Aptamers are often susceptible to rapid degradation. AptaBLE integrates proprietary databases and insights into nucleic acid self-structures to guide the design of aptamers with superior nuclease resistance—achieved without chemical modifications. This ensures that aptamers remain stable and functional in challenging biological environments.
Off-target interactions pose significant risks in therapeutic and diagnostic applications. AptaBLE employs AI models to counter-select aptamers against the human proteome, a task virtually impossible using traditional methods. Aptamers refined through this process demonstrate minimal off-target binding, ensuring greater safety and specificity.
In summary, with AptaBLE, we empower researchers to overcome the bottlenecks of traditional aptamer discovery, delivering precise, stable, and cost-effective solutions for targeting challenges in days—not months.
PreprintRead the AptaBLE preprint on bioRxiv