Inspired by the success of large language models (LLMs) in natural language processing, cell language models (CLMs) have emerged as a promising paradigm to learn cell representations from high-dimensional single-cell data—particularly transcriptomic profiles from scRNA-seq. These foundation models have shown remarkable potential across a variety of downstream applications. However, there remains a lack of foundation models for scATAC-seq data, which measures chromatin accessibility at single-cell level and is critical for decoding epigenetic regulation. Developing such model is considerably more challenging due to the unique characteristics of scATAC-seq data, including the vast number of chromatin regions, lack of standardized annotations, extreme sparsity, and near-binary distributions. To address these challenges, we systematically explore various strategies and propose CLM-Access, a specialized foundation model for scATAC-seq data. CLM-Access incorporates three main innovations: (1) an unified data processing pipeline that maps 2.8 million cells onto an unified reference of over 1 million chromatin regions; (2) a specialized patching and embedding strategy to effectively manage high-dimensional inputs; and (3) a tailored masking and loss function design that preserves fine-grained regional information while enhancing training efficiency and representation quality. With comprehensive benchmarks, we show that CLM-Access significantly outperforms existing methods in key downstream tasks, including batch effect correction, cell type annotation, RNA expression prediction, and multi-modal integration. This work establishes a scalable and interpretable foundation model for single-cell epigenomic analysis and expands the application of CLMs in single-cell research.

Multiple instance learning (MIL) has been successfully applied for whole slide images (WSIs) analysis in computational pathology, enabling a wide range of prediction tasks from tumor subtyping to inferring genetic mutations and multi-omics biomarkers. However, existing MIL methods predominantly focus on single-task learning, resulting in not only overall low efficiency but also the overlook of inter-task relatedness. To address these issues, we proposed an adapted architecture of Multi-gate Mixture-of-experts with Multi-proxy for Multiple instance learning (M4), and applied this framework for simultaneous prediction of multiple genetic mutations from WSIs. The proposed M4 model has two main innovations: (1) adopting a multi-gate mixture-of-experts strategy for multiple genetic mutation simultaneous prediction on a single WSI; (2) introducing a multi-proxy CNN construction on the expert and gate networks to effectively and efficiently capture patch-patch interactions from WSI. Our model achieved significant improvements across five tested TCGA datasets in comparison to current state-of-the-art single-task methods. The code is available at: https://github.com/Bigyehahaha/M4.

The process of drug discovery is widely known to be lengthy and resource-intensive. Artificial Intelligence approaches bring hope for accelerating the identification of molecules with the necessary properties for drug development. Drug-likeness assessment is crucial for the virtual screening of candidate drugs. However, traditional methods like Quantitative Estimation of Drug-likeness (QED) struggle to distinguish between drug and non-drug molecules accurately. Additionally, some deep learning-based binary classification models heavily rely on selecting training negative sets. To address these challenges, we introduce a novel unsupervised learning framework called DrugMetric, an innovative framework for quantitatively assessing drug-likeness based on the chemical space distance. DrugMetric blends the powerful learning ability of variational autoencoders with the discriminative ability of the Gaussian Mixture Model. This synergy enables DrugMetric to identify significant differences in drug-likeness across different datasets effectively. Moreover, DrugMetric incorporates principles of ensemble learning to enhance its predictive capabilities. Upon testing over a variety of tasks and datasets, DrugMetric consistently showcases superior scoring and classification performance. It excels in quantifying drug-likeness and accurately distinguishing candidate drugs from non-drugs, surpassing traditional methods including QED. This work highlights DrugMetric as a practical tool for drug-likeness scoring, facilitating the acceleration of virtual drug screening, and has potential applications in other biochemical fields.

Although substantial efforts have been made using graph neural networks (GNNs) for artificial intelligence (AI)-driven drug discovery, effective molecular representation learning remains an open challenge, especially in the case of insufficient labeled molecules. Recent studies suggest that big GNN models pre-trained by self-supervised learning on unlabeled datasets enable better transfer performance in downstream molecular property prediction tasks. However, the approaches in these studies require multiple complex self-supervised tasks and large-scale datasets , which are time-consuming, computationally expensive and difficult to pre-train end-to-end. Here, we design a simple yet effective self-supervised strategy to simultaneously learn local and global information about molecules, and further propose a novel bi-branch masked graph transformer autoencoder (BatmanNet) to learn molecular representations. BatmanNet features two tailored complementary and asymmetric graph autoencoders to reconstruct the missing nodes and edges, respectively, from a masked molecular graph. With this design, BatmanNet can effectively capture the underlying structure and semantic information of molecules, thus improving the performance of molecular representation. BatmanNet achieves state-of-the-art results for multiple drug discovery tasks, including molecular properties prediction, drug-drug interaction and drug-target interaction, on 13 benchmark datasets, demonstrating its great potential and superiority in molecular representation learning.