Applying Bioinformatics to Research in Immune, Muscle, and Bone Diseases

National Institute Of Arthritis And Musculoskeletal And Skin Diseases

Investigators

Linked publications, trials & patents

Abstract

The Biodata Mining and Discovery Section projects are listed below, including major accomplishments: - Investigation of genetic causes for severe calcinosis through whole genome sequencing Processed 56 Whole Genome Sequencing (WGS) samples from patients with sporadic or familial severe calcinosis. Performed data Quality Control (QC) and initial analysis. Provided consultations on the analysis of short-tandem repeats, mitochondria mutations, HLA typing and somatic mutations. - Investigation of genetic causes for Systemic Lupus Erythematous through whole genome sequencing Processed more than 230 new WGS samples sequenced by the NIAMS core. Implemented new sample QC tool that detects and prevents sample contamination. Performed trio-based WGS mutation analysis. - Family-based mutation analysis in the NIAMS Whole Exome Sequencing (WES) cohort Performed trio-based mutation analysis in more than 15 NIAMS patients with various diseases including Systemic Juvenile Idiopathic Arthritis (sJIA), Adult-Onset Stills Disease (AOSD), Takayasus arteritis, Relapsing Polychondritis, and Behcets-like disease. Identified a disease-causing mutation in the RELA gene. - Population-based mutation analysis in the NIAMS WES cohort Performed cohort analysis in non-trio samples for patients with Takayasus arteritis, Relapsing Polychondritis and sJIA. Performed joint variant call on more than 3000 WES Relapsing Polychondritis and control samples. Performed joint variant call on more than 3000 WES Takayasus arteritis and control samples. Provided guidance on variant enrichment analysis for the cohort. - The role of a NEMO isoform in the newly characterized autoinflammatory disease, NDAS A higher proportion of alternative splicing was identified in multiple patients with an autoinflammatory disease that are associated with an IKBKG mutation, leading to increased IFN and NF-kB responses. It was determined that patients with this mutation express a truncated NEMO isoform protein and exhibit a pediatric autoinflammatory syndrome, newly defined as NDAS. - Exploring human neutrophil subsets using single cell RNA-Seq Neutrophils are low mRNA cells that are difficult to study using typical single cell RNA-Seq workflows. Alternative computational methods were explored, and improved workflows were established, which were applied to scRNA-Seq studies of neutrophils from whole blood, leading to the identification of novel subsets of neutrophils with distinct biological functions. - Single cell RNA-Seq on whole blood of baracitinib treated Juvenile Dermatomyostis (JDM) patients to reveal disease signature and effect of treatment Single cell RNA-Seq was performed to characterize the transcriptomes of JDM patients before treatment and at 3 timepoints after treatment with baracitinib. By comparing JDM patients before treatment with healthy controls, a cell specific signature of JDM disease has been explored, along with transcriptomic changes induced by the treatment. - Role of U12 splicing genes in Lupus patients U12 splicing genes were identified and their gene expression profiles were characterized in Lupus patients. A specific single cell RNA-seq data analysis pipeline was implemented to further characterize U12 genes to identify novel cell types and their markers in Lupus patients. A scoring system has been established to measure the expression profiles of customized gene-sets. -Pathway enrichment analysis on genes correlated to UDP-glucose ceramide glucosyltransferase (UGCG) in tumor infiltrating lymphocytes (TILs) The public GEO mouse TIL samples were identified and their RNA-seq gene expression data downloaded from ARCHS4. Data were analyzed using R and Partek GS. Genes with a UGCG correlation > 0.80 were used for pathway enrichment analysis with Enrichr, assisted by several in-house developed R scripts for results retrieval, organization, and graphical data analysis. Interferon-gamma signaling pathway was identified as the most significantly enriched, supporting the hypothesis that UGCG plays a previously unrecognized important role in NK cell immunometabolism. - Effect of Stat4 deletion on differentiation of large intestine ILC1s compared to Natural Killer cells in inflamed large intestine - 3D genomic organization of Mdm1-Il22-Ifng locus - Regulation of skeletal muscle stem cell function by EZH1 - Exploration of lineage specification through epigenetic regulation of Pax7 - Functions of Psat1 in muscle stem cell activation - Development of a scRNA-seq pipeline to analyze circulating human neutrophils - Investigaton of the functions of oral epithelial transcription factor Pitx1 in cutaneous wounds repair - Spatial transcriptome analysis of inflammatory mouse kidney - Investigation of the internalization of neutrophil extracellular trap-bound RNA by endothelial cells - Transcriptional programing of helper T cell metabolism by the IL-2-STAT5 axis - Essential role of glycosphingolipid metabolism for natural killer cell development and function - Gene expression profiles of cancer cells treated with copper and disulfiram - Transcriptional regulation of jaw osteoblasts - Rare variant analysis in systemic Juvenile Idiopathic Arthritis (sJIA) - Investigation of genome-wide modifications from CRISPR editing through whole genome sequencing - Mutation analysis in patients with early onset Merkel cell carcinoma Additional major accomplishments of the BMDS this year include: -Visualization of HiC data with HiCExplorer Multiple HiCExplorer tools were applied to visualize the hiC data. The contact matrices files were converted to h5 format with hicConvertFormat, normalized with smallest model with hicNormaliz, and corrected with hicCorrectMatrix. Then, hicFindTADs was applied to the corrected normalized matrices to calculate a genome-wide TAD separation score. In the end, domains, loops, and TAD score data were plotted with hicPlotTADs over regions of interest. -CUT&Tag analysis Cut&Tag data processing and analysis pipeline originally developed in the Henicoff lab has been customized and implemented for data from wild type or doxycycline-treated mouse keratinocytes. After pre-processing, the reads are aligned to the mm10 genome with bowtie2. Both SEACR and MACS2 are used for the peak calling. Counts from bam files are collected and used as input to DESeq2 for differentially expressed genes (DEG) analysis. -Single cell multiome analysis Seurat multimodal pipeline has been customized and applied to multi-modal experiments generated with differentiated mouse muscle cells. The multi-omic dataset was realigned to mm10. First, both RNA-seq count matrix and the ATAC-seq peak matrix were cleaned. Next, the Weighted Nearest Neighbor (WNN) graph was constructed to learn the relative utility of each data modality in each cell. The R packages Seurat and Signac were used for data scaling, transformation, clustering, dimensionality reduction, differential expression analysis, and most visualizations. The R package Monocle3 was used for trajectory analysis. -Built a more robust RNA-seq data analysis pipeline This pipeline uses STAR (instead of TopHat) for mapping bulk RNA reads and DESeq2 for downstream analysis -Built an R package for pathway and gene set enrichment analysis Based on our previous work, a new R package (richPathR) has been developed for pathway and gene set enrichment analysis. Powerful yet user-friendly, it is based on the popular Enrichr pathway analysis web tool, making multiple gene lists-based analysis possible. The package offers multiple utilities for enrichment data manipulation and visualization.