Prediction of aptamer affinity using an artificial intelligence approach.

Area of Science:

• Biotechnology
• Computational Biology
• Drug Discovery

Background:

• Aptamers are oligonucleotide molecules that bind specific targets, analogous to antibodies.
• Traditional aptamer selection via SELEX is often time-consuming and challenging.
• Computational methods are increasingly vital for aptamer design and prediction.

Purpose of the Study:

• To review advancements in AI and computational strategies for predicting aptamer binding ability.
• To explore the integration of machine learning, deep learning, and structure-based methods for aptamer discovery.

Main Methods:

• Systematic Evolution of Ligands by Exponential Enrichment (SELEX) and its limitations.
• Structure-based computational approaches for RNA and DNA secondary and 3D structure prediction.
• Molecular docking and molecular dynamics simulations for aptamer-target interactions.
• Application of Artificial Intelligence (AI), machine learning (ML), and deep learning (DL) models for binding prediction.

Main Results:

• AI and DL models show promise in accurately predicting aptamer-target binding properties.
• Computational methods, including structure-based design and simulations, aid in aptamer selection.
• Integration of diverse computational strategies can overcome limitations of experimental aptamer identification.

Conclusions:

• AI-driven pipelines offer a precise and reliable method for forecasting aptamer binding.
• Advanced computational techniques are essential for efficient and effective aptamer discovery and development.
• Future research should focus on refining AI models and integrating them with experimental validation for robust aptamer selection.