Biology subjects

Hui, D.

Publications and source records attributed to Hui, D..

5 recordsLinked to original sources

bioRxiv2025

Generative genomics accurately predicts future experimental results

Realizing AIs promise to accelerate biomedical research requires AI models that are both accurate and sufficiently flexible to capture the diversity of real-life experiments. Here, we describe a generative genomics framework for AI-based experimental prediction that mirrors the process of designing and conducting an experiment in the lab or clinic. We created GEM-1 (Generate Expression Model-1), an AI system that effectively models the enormous range of bulk and single-cell gene expression experiments performed by scientists and benchmarked its performance across multiple biological axes. GEM-1s prediction of future gene expression experiments-RNA-seq data deposited in public archives after our training data cutoff-yielded accuracy comparable to the best-possible performance estimated by comparing the results of matched lab experiments. Overall, our approach illustrates the transformative potential of generative genomics for applications ranging from predicting cellular perturbations in vitro to de novo generation of data from large clinical cohorts.

genomics↗

bioRxiv2022

Cell type- and state- resolved immune transcriptomic profiling identifies glucocorticoid-responsive molecular defects in multiple sclerosis T cells

The polygenic and multi-cellular nature of multiple sclerosis (MS) immunopathology necessitates cell-type-specific molecular studies in order to improve our understanding of the diverse mechanisms underlying immune cell dysfunction in MS. Here, by generating a dataset of 1,075 transcriptomes from 209 participants (167 MS and 42 healthy), we assessed MS-associated transcriptional changes in six implicated cell-type-states: naive and memory helper T cells and classical monocytes purified from peripheral blood, each in their primary (ex vivo, unstimulated) and in vitro stimulated states. Our data suggest that primary profiles show larger MS-associated differences than the post-stimulation contexts. We further identified shared and distinct changes in individual genes, biological pathways, and co-expressed gene modules in MS T cells and monocytes, and prioritized genes such as ZBTB16 as MS-associated regulators in both cell types. Of six identified MS-associated co-expressed gene modules, three (two lymphoid and one myeloid) were replicated in independent data from peripheral blood mononuclear cells (PBMC) and monocyte-derived macrophages. A subsequent in silico drug screen prioritized small-molecule compounds for reversing the perturbation of the MS-associated modules. The effects of glucocorticoid receptor agonists as the top-identified therapeutic class for the replicated T cell modules were validated using targeted in silico analyses and in vitro experiments, suggesting the coordinated dysregulation of glucocorticoid-responsive genes in MS T cells. In summary, our study identifies and validates individual genes and co-expressed gene modules from T and myeloid cells that are perturbed in MS, offering new targets for therapeutic discovery and biomarker development to guide the management of MS.

genomics↗

bioRxiv2022

Resolving clonal substructure from single cell genomic data using CopyKit

High-throughput methods for single cell copy number sequencing have enabled the profiling of thousands of cells in parallel, yet there remains a significant bottleneck for data analysis. Here we present CopyKit, a comprehensive set of computational methods for the pre-processing and analysis of single cell copy number data to resolve clonal substructure and reconstruct genetic lineages in tumors. We performed single cell DNA sequencing of 2977 cells from multiple spatial regions in two liver metastasis and 7365 cells from three primary tumors with matched metastatic tissues. In the liver metastases, CopyKit resolved clonal substructure in different spatial regions, which revealed both clonal intermixing and spatial segregation in the tumor mass. In the matched metastatic colorectal and breast cancers, CopyKit resolved metastatic lineages and identified subclones and genomic events that were associated with metastases. These applications show that CopyKit is comprehensive tool for resolving copy number substructure in tumors.

genomics↗

bioRxiv2020

The RNA phosphatase PIR-1 regulates endogenous small RNA pathways in C. elegans

Eukaryotic cells regulate 5 triphosphorylated (ppp-) RNAs to promote cellular functions and prevent recognition by antiviral RNA sensors. For example, RNA capping enzymes possess triphosphatase domains that remove the {gamma} phosphates of ppp-RNAs during RNA capping. Members of the closely related PIR1 family of RNA polyphosphatases remove both the {beta} and {gamma} phosphates from ppp-RNAs. Here we show that C. elegans PIR-1 dephosphorylates ppp-RNAs made by cellular RdRPs and is required for the maturation of 26G-RNAs, Dicer-dependent small RNAs that regulate thousands of genes during spermatogenesis and embryogenesis. PIR-1 also regulates the CSR-1 22G-RNA pathway and has critical functions in both somatic and germline development. Our findings suggest that PIR-1 modulates both Dicer-dependent and - independent Argonaute pathways, and provide insight into how cells and viruses use a conserved RNA phosphatase to regulate and respond to ppp-RNA species.

molecular biology↗

bioRxiv2019

One-time Nitrogen Fertilization Shifts Switchgrass Soil Microbiomes within a Context of Larger Spatial and Temporal Variation

Soil microbiome responses to short-term nitrogen (N) inputs within the context of existing spatio-temporal variability remain uncertain. Here, we examined soil bacterial and fungal communities pre/post-N fertilization in an 8 year-old switchgrass field, in which twenty-four plots received N fertilization at three levels (0, 100, and 200 kg N ha-1 as NH4NO3) for the first time since planting. Soils were collected at two depths, 0-5 and 5-15 cm, for DNA extraction and amplicon sequencing of 16S rRNA genes and ITS regions, and soil metagenomic analysis. Baseline assessment prior to fertilization revealed no pre-existing differences in either bacterial or fungal communities across plots. The one-time N fertilization increased switchgrass yields and tissue N content, and the added N was nearly completely removed from the soil of fertilized plots by the end of the growing season. Both bacterial/archaeal and fungal communities showed large spatial (by depth) and temporal variation (by season) within each plot, accounting for 17 and 12-22 % of the variation in bacterial/archaeal and fungal community composition, respectively. While N fertilization effects accounted for only ~4% of overall variation, some specific microbial groups, including the bacterial genus Pseudonocardia and the fungal genus Archaeorhizomyces, were notably repressed by fertilization at 200 kg N ha-1. Bacterial groups varied with both depth in the soil profile and time of sampling, while temporal variability shaped the fungal community more significantly than vertical heterogeneity in the soil. Thus, variability within the field might override the changes induced by N addition. Continued analyses of these trends over time with fertilization and management are needed to understand whether these transient effects change over time.

ecology↗