Core Viewpoint - The article discusses the development of a generative AI framework called GRAPE-LM for the efficient evolution of RNA aptamers, addressing the challenges of traditional labor-intensive screening methods in aptamer discovery [2][3]. Group 1: RNA Aptamer Evolution - RNA aptamers are short single-stranded DNA or RNA oligonucleotides that can fold into specific three-dimensional structures, allowing them to bind proteins or small molecules with high affinity, thus holding potential for drug development and diagnostic probes [5]. - Traditional methods for aptamer discovery, such as SELEX, require multiple rounds of labor-intensive screening and often yield aptamers with low affinity and specificity, limiting their potential applications [5][6]. Group 2: GRAPE-LM Framework - The GRAPE-LM framework combines a Transformer-based conditional autoencoder with a nucleic acid language model, guided by CRISPR-Cas-based aptamer screening data from the intracellular environment [3][7]. - This framework allows for a single-round evolution of RNA aptamers, significantly improving efficiency compared to traditional SELEX methods, which typically require 6-16 rounds to achieve similar results [8]. Group 3: Advantages of CRISmers and GRAPE-LM - The CRISmers system enhances the biological relevance of aptamer screening by conducting selections in the intracellular environment, capturing endogenous biological mechanisms that are often missed in traditional methods [6][9]. - The combination of CRISmers and GRAPE-LM allows for a reduction in the initial library size needed for effective screening, requiring only about 10^8 sequences compared to the 10^14-10^16 required by classical SELEX methods [8]. Group 4: Future Prospects - The integration of physical models alongside language models in future iterations of GRAPE-LM could further enhance its generative capabilities, with potential applications already demonstrated in the efficient discovery of peptide molecules [12]. - The work has received positive feedback from experts in the field, highlighting its innovative approach and potential impact on RNA aptamer research [12].