Designing biologically plausible regulatory sequences, such as enhancers, is a key challenge in synthetic biology. Current approaches either classify active enhancers or generate synthetic sequences independently but often fail to address the need for cell-type-specific recruitment of multiple transcription factors (TFs) for effective transcription activation. This oversight leads to suboptimal designs. To overcome these limitations, we propose Leonine, a novel framework redefining enhancer sequence design as a multimodal question-answering task within a cellular context. Leonine utilizes large language models (LLMs) trained on DNA sequences and integrates multimodal cellular data-including promoter sequences, gene expression profiles, and cell type information-to optimize enhancer sequences for specific regulatory environments. A large-scale dataset of over 1.5 million cell-type-specific promoter-enhancer pairs and a robust evaluation benchmark were developed to support this effort. Extensive evaluations across seven cell types demonstrate that Leonine outperforms state-of-the-art LLMs, generating biologically and functionally coherent sequences. This work introduces a new paradigm for context-aware DNA sequence design, advancing research in synthetic biology and gene regulation.

The single-cell sequencing revolution enables simultaneous molecular profiling of various modalities across thousands of individual cells, allowing scientists to investigate the diverse functions of complex tissues. Among all the analysis steps, assigning individual cells to specific types is fundamental for understanding cellular heterogeneity. However, this process is labor-intensive and requires extensive expert knowledge. Recent advances in large language models (LLMs) have demonstrated their ability to automatically extract biological knowledge, such as marker genes, promoting efficient, and automated cell-type annotations. To evaluate the capability of modern LLMs in automating the cell-type identification process, we first introduce an automated cell-type annotation method with comprehensive benchmark: Single-cell Omics Arena). Specifically, we began by compiling 11 publicly available single-cell RNA sequencing (scRNA-seq) datasets and evaluating eight LLMs across 1226 cell-type annotation-related tasks. This effort established a foundation for automated cell-type annotation from scRNA-seq data using interpretable features such as gene names. Building upon this benchmark, we introduced domain-specific chain-of-thought prompting techniques to enhance the accuracy of cell-type annotation and facilitate the extraction of relevant biological insights. Finally, to accommodate non-interpretable features, we proposed to leverage a pretrained VAE-based cross-modality translation module to convert features such as epigenetic marks into interpretable representations, which enables the seamless extension of LLM-based cell-type annotation to non-RNA-based sequencing technologies. In summary, our benchmark provides key insights into automated cell-type annotation from scRNA-seq data and demonstrates the potential of cross-modality translation for handling non-interpretable features.

MOTIVATION: Single-cell clustering plays a crucial role in distinguishing between cell types, facilitating the analysis of cell heterogeneity mechanisms. While many existing clustering methods rely solely on gene expression data obtained from single-cell RNA sequencing techniques to identify cell clusters, the information contained in mono-omic data is often limited, leading to suboptimal clustering performance. The emergence of single-cell multi-omics sequencing technologies enables the integration of multiple omics data for identifying cell clusters, but how to integrate different omics data effectively remains challenging. In addition, designing a clustering method that performs well across various types of multi-omics data poses a persistent challenge due to the data's inherent characteristics. RESULTS: In this paper, we propose a graph-regularized multi-view ensemble clustering (GRMEC-SC) model for single-cell clustering. Our proposed approach can adaptively integrate multiple omics data and leverage insights from multiple base clustering results. We extensively evaluate our method on five multi-omics datasets through a series of rigorous experiments. The results of these experiments demonstrate that our GRMEC-SC model achieves competitive performance across diverse multi-omics datasets with varying characteristics. AVAILABILITY AND IMPLEMENTATION: Implementation of GRMEC-SC, along with examples, can be found on the GitHub repository: https://github.com/polarisChen/GRMEC-SC.