Explore the vast range of titles available.

RNA plays a crucial role not only in information transfer as messenger RNA during gene expression but also in various biological functions as non-coding RNAs. Understanding mechanical mechanisms of function needs tertiary structure information; however, experimental determination of three-dimensional RNA structures is costly and time-consuming, leading to a substantial gap between RNA sequence and structural data. To address this challenge, we developed NuFold, a novel computational approach that leverages state-of-the-art deep learning architecture to accurately predict RNA tertiary structures. NuFold is a deep neural network trained end-to-end for the output structure from the input sequence. NuFold incorporates a nucleobase center representation, which enables flexible conformation of ribose rings. Benchmark study showed that NuFold clearly outperformed energy-based methods and demonstrated comparable results with existing state-of-the-art deep-learning-based methods. NuFold exhibited a particular advantage in building correct local geometries of RNA. Analyses of individual components in the NuFold pipeline indicated that the performance improved by utilizing metagenome sequences for multiple sequence alignment and increasing the number of recycling. NuFold is also capable of predicting multimer complex structures of RNA by linking the input sequences. The 3D structure of RNA is crucial for unraveling its functions. Here, the authors developed NuFold, an end-to-end deep learning-based approach that predicts all-atom RNA 3D structures from sequences. NuFold’s all-atom representation is designed to accurately represent the high flexibility of RNA.

Cryo-electron microscopy (cryo-EM) has advanced structural biology by enabling the determination of numerous macromolecular structures. However, accurate interpretation of cryo-EM maps, particularly at medium to low resolutions, requires sophisticated computational tools to build precise atomic models. To address this challenge, we present the EMSuite Server (https://em.kiharalab.org, Fig. 1), an integrated web-based platform developed by the Kihara Lab. EMSuite offers a suite of 14 state-of-the-art algorithms for cryo-EM structure modeling, validation, and refinement. These include tools such as DeepMainmast and CryoREAD, which use deep learning for de novo protein and nucleic acid structure modeling. For 5-10 Å resolution maps, the server provides DiffModeler and DMcloud that perform model-map fitting based on diffusion models. For model quality assessment and refinement, the server provides DAQ-Score and DAQ-Refine. Figure 2 shows examples computed by DeepMainmast and DAQ score using the EMSuite server. As shown in Figure 2A, DeepMainmast accurately built a protein structure model from the cryo-EM map, EMD-1461 (resolution 3.8 Å), achieving an RMSD of 1.30 Å and a TM-score of 0.97 (Fig. 2A). Additionally, DAQ-Score was used to evaluate the local model quality of PDB structure 7fet fitted into EMD-32563. Figure 2B shows color-coded DAQ(AA) scores mapped onto the PDB structure, while Figure 2C presents the plots of the DAQ(AA) scores along the sequence positions. A region around residues 590–620 of chain B indicated negative DAQ scores, suggesting a possible misalignment in the PDB structure model. These visualizations allow users to identify and correct problematic regions in fitted models. This presentation highlights the EMSuite Server’s key functionalities, including main algorithms such as DeepMainmast, CryoREAD, and the newly released DMcloud. The server integrates deep learning and diffusion-based methods to provide a comprehensive and user-friendly platform for cryo-EM structure modeling, validation, and refinement. We demonstrate how the EMSuite server enables the generation of accurate atomic models from cryo-EM maps. By leveraging these tools, researchers can build high-quality structural models that enhance the understanding of molecular functions and mechanisms.

The rapid expansion of biomedical literature has made comprehensive manual synthesis increasingly difficult to perform effectively, creating a pressing need for AI systems capable of reasoning across verified evidence rather than merely retrieving it. However, existing retrieval- augmented generation (RAG) methods often fall short when faced with complex biomedical questions that require iterative reasoning and multi-step synthesis. Here, we developed Queryome, a deep research system consisting of specialized large language model (LLM) agents that can adapt their orchestration dynamically to a wide range of queries. Using a hybrid semantic-lexical retrieval engine spanning 28.3 million PubMed abstracts, it performs iterative, evidence-grounded synthesis. On the MIRAGE benchmark, Queryome achieved 88.98 % accuracy, surpassing prior systems by up to 14 points, and improved reasoning accuracy on the biomedical Human's Last Exam (HLE) subset from 15.8% to 19.3%. Moreover, in a task for constructing a review article, it earned the highest composite score in comparison with Deep Research from OpenAI, Google, Perplexity, and Scite.AI, reflecting its strong literature retrieval and synthesis capabilities.

Biomolecules rely on water and ions for stable folding, but these interactions are often transient, dynamic, or disordered and thus hidden from experiments and evaluation challenges that represent biomolecules as single, ordered structures. Here, we compare blindly predicted ensembles of water and ion structure to the cryo-EM densities observed around the ribozyme at 2.2-2.3 Å resolution, collected through target R1260 in the CASP16 competition. 26 groups participated in this solvation 'cryo-ensemble' prediction challenge, submitting over 350 million atoms in total, offering the first opportunity to compare blind predictions of dynamic solvent shell ensembles to cryo-EM density. Predicted atomic ensembles were converted to density through local alignment and these densities were compared to the cryo-EM densities using Pearson correlation, Spearman correlation, mutual information, and precision-recall curves. These predictions show that an ensemble representation is able to capture information of transient or dynamic water and ions better than traditional atomic models, but there remains a large accuracy gap to the performance ceiling set by experimental uncertainty. Overall, molecular dynamics approaches best matched the cryo-EM density, with blind predictions from bussilab_plain_md, SoutheRNA, bussilab_replex, coogs2, and coogs3 outperforming the baseline molecular dynamics prediction. This study indicates that simulations of water and ions can be quantitatively evaluated with cryo-EM maps. We propose that further community-wide blind challenges can drive and evaluate progress in modeling water, ions and other previously hidden components of biomolecular systems.