Qinhu Zhang

Deciphering the relationships between cis-regulatory elements (CREs) and target gene expression has long been a challenging problem in molecular biology. However, predicting gene expression from hundreds of candidate cis-regulatory elements (cCREs) requires models that scale to long, noisy inputs while retaining interpretable regulatory structure. Existing Transformer-based approaches typically attend over all nucleotides and all surrounding cCREs, diluting causal signals when hundreds of elements compete for limited model capacity. Here we introduce a two-stage selective framework (TSSF) that performs hierarchical refinements: nucleotide-level masking within each cCRE, followed by cCRE-level selection around each gene, implemented with information-bottleneck priors and a fully Transformer-based architecture. Across 70 human cell types and tissues, TSSF and lightweight variants improve expression prediction and enhancer-gene prioritization relative to strong baselines, including on cross-cell-line and cell-type-specific benchmarks. Prediction-stratified analysis motivates a distance-decay prior that aligns attention with long-range regulatory geometry, and chromatin-contact augmentation improves recovery of distal links. Motif analyses of high-confidence predictions recover proximal and distal regulatory programs, supporting mechanistic interpretability. TSSF offers a general strategy for scalable, interpretable modeling of high-dimensional regulatory inputs in genomics.

Deciphering the relationships between cis -regulatory elements (CREs) and target gene expression has been a long-standing unsolved problem in molecular biology, and the dynamics of CREs in different cell types make this problem more challenging. To address this challenge, we propose a sc alable computational framework for p redicting g ene e xpression (ScPGE) from discrete candidate CREs (cCREs). ScPGE assembles DNA sequences, transcription factor (TF) binding scores, and epigenomic tracks from discrete cCREs into three-dimensional tensors, and then models the relationships between cCREs and genes by combining convolutional neural networks with transformers. Compared with current state-of-the-art models, ScPGE exhibits superior performance in predicting gene expression and yields higher accuracy in identifying active enhancer–gene interactions through attention mechanisms. By comprehensively analyzing ScPGE's predictions, we find a pattern in true positives (TPs) that the regulatory effect of cCREs on genes decreases with distance. Inspired by the pattern, we design two methods to enhance the ability to capture distal cCRE–gene interactions by incorporating chromatin loops into the ScPGE model. Furthermore, ScPGE accurately discovers some crucial TF motifs within prioritized cCREs and reveals the different regulatory types of these cCREs.

Deciphering the relationships between cis-regulatory elements (CREs) and target gene expression has been a long-standing unsolved problem in molecular biology, and the dynamics of CREs in different cell types make this problem more challenging. To address this challenge, we propose a sc alable computational framework for p redicting g ene e xpression (ScPGE) from discrete candidate CREs (cCREs). ScPGE assembles DNA sequences, transcription factor (TF) binding scores, and epigenomic tracks from discrete cCREs into 3-dimensional tensors, and then models the relationships between cCREs and genes by combining convolutional neural network with transformer. Compared with current state-of-the-art models, ScPGE exhibits superior performance in predicting gene expression and yields higher accuracy in identifying active enhancer-gene interactions through attention mechanisms. By comprehensively analyzing ScPGE’s predictions, we find a pattern in true positives (TPs) that the regulatory effect of cCREs on genes decreases with distance. Inspired by the pattern, we design two methods to enhance the ability to capture distal cCRE-gene interactions by incorporating chromatin loops into the ScPGE model. Furthermore, ScPGE accurately discovers some crucial TF motifs within prioritized cCREs and reveals the different regulatory types of these cCREs.

Single-cell RNA sequencing (scRNA-seq) technology has opened new frontiers in biomedical research, offering insights into cellular heterogeneity. Accurate cell clustering and batch effect correction are essential in single-cell RNA sequencing (scRNA-seq) data analysis, forming the foundation for downstream steps. However, most methods handle these tasks separately, limiting their applicability across diverse datasets. To address these challenges, we introduce Relational Graph Convolutional Network with Batch Awareness (RGCN-BA), a deep learning framework that integrates cell clustering and batch effect correction into a unified model. For multi-batch datasets, RGCN-BA leverages relational graph convolutional network to process batch information as distinct edge types, followed by a batch correction layer for global alignment. For single-batch data, it functions with a single edge type. Experiments on both multi-batch and single-batch datasets demonstrate that RGCN-BA outperforms both specialized clustering methods and batch effect correction methods. This versatility in handling both tasks positions RGCN-BA as a powerful tool for enhancing scRNA-seq data analysis.