Biology subjects

Xu, A. G.

Publications and source records attributed to Xu, A. G..

3 recordsLinked to original sources

bioRxiv2026

INS-fMRI: a novel method for mapping mesoscale connectome networks in vivo in nonhuman primates

Mapping brain connections is paramount in neuroscience. Lacking has been a method that can systematically map mesoscale (<1mm) connections in the brain in vivo at brainwide scale. Mesoscale mapping is particularly important for NHPs in which information is represented in submillimeter clusters (e.g. cortical columns, striosomes, thalamic rods). Here, we present a method called INS-fMRI, which comprises optical stimulation of single sites in the brain using pulsed Infrared Neural Stimulation (INS), coupled with mapping of functionally connected sites in the brain using ultrahigh field fMRI. It is the only method available for mapping mesoscale connections rapidly (multiple networks mapped within single MRI sessions), systematically (allowing comparison of networks within single individuals), in vivo (no animal sacrifice needed), at whole brain scale, and without viral transfection. This method has broad applicability to a wide range of neuroscience questions. It can be used at any brain site, on different animal models, and on different MRI platforms. The ability to probe circuits repeatedly within single individuals also opens doors for studying changes in mesoscale circuits over time (e.g. for development, aging, disease studies). In addition, as this method circumvents the need for viral transfection, there is exciting potential for human clinical application.

neuroscience↗

bioRxiv2025

A SNP Foundation Model: Application in Whole-Genome Haplotype Phasing and Genotype Imputation

Millions of human genomes have been genotyped by national biobanks worldwide. Training large language models (LLM) with this data may lead to a universal model of human genome with tremendous potential. Yet the quadrillions (1015) of nucleotides-- resulting from genome length multiplied by population size--pose formidable challenges for modeling. In this study, we propose a novel AI framework designed to scale with this data and support diverse analytical tasks. To demonstrate this scheme, we developed SNPBag--a foundation model focusing on single nucleotide polymorphism (SNP). With 0.8 billion parameters, it is trained on one million synthesized human genomes, corresponding to a total of 6 trillion SNP tokens. SNPBag showed superior performance in benchmarking of multiple tasks. In genotype imputation, it achieves state-of-the-art (SOTA) accuracy. In haplotype phasing, it rivals the best method with a 72-fold speedup. By encoding 6 million SNPs per genome into a 0.75 MB embedding, SNPBag enables efficient storage, transfer and downstream applications. In particular, the genome embeddings facilitate rapid ancestry inference across global populations and detection of genetic relationships up to 12th-degree relatives. Collectively, SNPBag introduces a new paradigm for scalable, unified and multitask analysis of the ever-growing human variation data.

genomics↗

bioRxiv2024

GeneBag: training a cell foundation model for broad-spectrum cancer diagnosis and prognosis with bulk RNA-seq data

Numerous Pre-trained cell foundation models (CFM) have been crafted to encapsulate the comprehensive gene-gene interaction network within cells, leveraging extensive single-cell sequencing data. These models have shown promise in various cell biology applications, including cell type annotation, perturbation inference, and cell state embedding, etc. However, their clinical utility, particularly in cancer diagnosis and prognosis, remains an open question. We introduce the GeneBag model, a novel CFM that represents a cell as "a bag of unordered genes" with continuous expression values and a full-length gene list. Pre-trained on single-cell data and fine-tuned on bulk RNA-seq datasets, GeneBag achieves superior performance across cancer diagnosis and prognosis scenarios. In a zero-shot learning setting, GeneBag can classify cancer and non-cancer tissues with approximately 96.2% accuracy. With fine-tuning, it can annotate 40 different types of cancers and corresponding normal biopsies with an overall accuracy of [~]97.2%. It notably excels in classifying challenging cancers such as bladder (93%) and stomach (90%). Furthermore, GeneBag is capable of cancer staging with 68.5% accuracy and 5-year survival prediction with an AUC of [~]80.4%. This study marks the first to demonstrate the potential of CFMs in RNA-based cancer diagnostics and prognostics, indicating a promising avenue for AI-assisted molecular diagnosis.

bioinformatics↗