Yaozhong Zhang

The paper introduces a structured-evidence framework for predicting bacteriophage-host interactions using a large language model, achieving high accuracy and enabling interpretable inference.

RiboPipe imputes codon-resolution Ribo-seq coverage for low-coverage transcripts using a unified framework that improves accuracy and efficiency.

This study evaluates complex features in drug-target interaction prediction and introduces a supervised vector quantization framework that improves performance and interpretability.

This paper introduces VQ-MIL, a weakly supervised vector quantization method for WSI classification, achieving state-of-the-art results on benchmark datasets.

PharaCon uses a conditional BERT framework to improve phage identification in metagenomic data by incorporating label constraints during pre-training.

The paper introduces scVQC, a supervised representation learning method for cell-type annotation, and shows it outperforms unsupervised and pre-trained foundation models in tissue-specific annotation.

This paper introduces STForte, a method for spatially resolved transcriptomics analysis that matches expression profiles with spatial topology, improving downstream analysis and spatial enhancement.

This paper introduces SCLSC, a supervised contrastive learning method for cell type annotation in scRNA-seq data, achieving better accuracy and scalability than existing methods.

The paper introduces a deep neural network using Gene Ontology to discover microbiome functionality in human diseases, finding novel important genes and functions across four conditions.

The paper analyzes how a pre-trained BERT model processes nucleotide sequences, identifies k-mer embeddings as key, and proposes an accelerated pre-training method.

The paper introduces INHERIT, a deep learning model that combines database-based and alignment-free methods for phage identification, achieving an F1-score of 0.9932 on a benchmark dataset.

This paper explores using refined BERT models for nanopore methylation detection, achieving competitive performance and faster inference speed compared to biRNN models.

RDBKE improves breakpoint resolution of read-depth SV callers using a deep segmentation model, increasing the number of SVs with more precise breakpoints on simulation and real data.

This paper introduces a base-calling method for nanopore sequencing by framing it as a one-dimensional segmentation task, achieving competitive accuracy compared to existing methods.