Single-round evolution of RNA aptamers with GRAPE-LM

2026-02-06 · Source: Machine learning : nature.com subject feeds · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Computational Biology · Depth: Expert, long

Summary

GRAPE-LM (generator of RNA aptamers powered by activity-guided evolution and language model) is a new generative AI framework designed for single-round RNA aptamer evolution, addressing the labor-intensive nature of traditional multi-round screening. This framework integrates a transformer-based conditional autoencoder with nucleic acid language models, guided by CRISPR-Cas-based aptamer screening data from intracellular environments. GRAPE-LM was validated on three distinct targets: the human T cell receptor CD3ε, the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein, and the human oncogenic transcription factor c-Myc. The system successfully generated RNA aptamers that outperformed those derived from multiple rounds of human selection and optimization, using only a single round of CRISPR-Cas-based screening data.

Key takeaway

For AI Researchers and Bioengineers focused on biomolecule discovery, GRAPE-LM offers a significant acceleration in RNA aptamer evolution. You should explore integrating similar generative AI frameworks with single-round experimental data to drastically reduce development timelines and resource expenditure for novel therapeutic or diagnostic aptamers. Consider leveraging the publicly available code and online platform for initial experimentation with the provided targets.

Key insights

GRAPE-LM enables single-round RNA aptamer evolution using AI, outperforming multi-round human methods.

Principles

• AI-driven generative models can accelerate biomolecule evolution.
• Activity-guided latent spaces enhance functional aptamer discovery.

Method

GRAPE-LM integrates a transformer-based conditional autoencoder with nucleic acid language models, guided by intracellular CRISPR-Cas screening data to generate RNA aptamers in one round.

In practice

• Utilize GRAPE-LM for rapid RNA aptamer discovery.
• Apply CRISPR-Cas screening data to guide generative AI models.

Topics

• RNA Aptamer Evolution
• Generative AI
• Nucleic Acid Language Models
• CRISPR-Cas Screening
• Biomolecule Design

Code references

• tansaox2008123/GRAPE-LM

Best for: AI Researcher, AI Scientist, Research Scientist

Related on AIssential

• Programmable RNA translation through deep learning-driven IRES discovery and de novo generation — An AI framework unifies IRES identification, optimization, and de novo generation for programmable RNA translation.
• Leap in DNA Synthesis Slashes Time to Build New Genetic Sequences — Sidewinder enables rapid, high-fidelity DNA synthesis, resolving a key bottleneck for generative AI in biology.
• Identification and engineering of highly functional potyviral proteases in cells using co-evolutionary models — Co-evolutionary models can accurately predict and engineer superior potyviral proteases for precise cellular control.
• [R] Genomic Large Language Models — Genomic LLMs can identify functional similarities in DNA beyond sequence alignment, indicating learned regulatory patterns.
• Orthrus: toward evolutionary and functional RNA foundation models — Orthrus is a Mamba-based RNA foundation model using contrastive learning and biological augmentations for superior RNA property prediction.
• Researchers try to cut the genetic code from 20 to 19 amino acids — Researchers from Columbia and Harvard successfully engineered *E.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine learning : nature.com subject feeds.