spatialDLPFC Poster GBD23

"Integrated single cell and unsupervised spatial transcriptomic analysis defines molecular anatomy of the human dorsolateral prefrontal cortex"

Poster for the Genome of Brain Disorders 2023

Pre-print: https://www.biorxiv.org/content/10.1101/2023.02.15.528722v1

Links from poster:
spatialDLPFC: https://research.libd.org/spatialDLPFC/
spatialLIBD: https://research.libd.org/spatialLIBD/
spatial registration tutorial: https://research.libd.org/spatialLIBD/articles/guide_to_spatial_registration.html

Transcript

1. Integrated single cell and unsupervised spatial transcriptomic analysis
   defines molecular anatomy of the human dorsolateral prefrontal cortex
   Louise A. Huuki-Myers1, Abby Spangler1, Nicholas J. Eagles1, Kelsey D. Montgomery1, Sang Ho Kwon1,2, Boyi Guo3, Melissa Grant-Peters4,5, Heena R. Divecha1, Madhavi
   Tippani1, Chaichontat Sriworarat1,2, Annie B. Nguyen1, Prashanthi Ravichandran6, Matthew N. Tran1, Arta Seyedian1, PsychENCODE consortium,Thomas M. Hyde1,7,8, Joel E.
   Kleinman1,7, Alexis Battle6,9,10,11 , Stephanie C. Page1, Mina Ryten4,5,12, Stephanie C. Hicks3,11, Keri Martinowich1,2,7, Leonardo Collado-Torres1*, Kristen R. Maynard1,2,7*
   1. Lieber Institute for Brain Development, 2. The Solomon H. Snyder Department of Neuroscience Johns Hopkins School of Medicine (JHSM), 3. Department of Biostatistics Johns Hopkins Bloomberg School of Public Health, 4. Genetics and Genomic Medicine Great Ormond
   Street Institute of Child Health University College London, 5. Aligning Science Across Parkinson’s Collaborative Research Network, 6. Department of Biomedical Engineering JHSM, 7. Department of Psychiatry and Behavioral Sciences JHSM, 8. Department of Neurology JHSM,
   9. Department of Computer Science JHU, 10. Department of Genetic Medicine JHSM, 11. Malone Center for Engineering in Healthcare JHU, 12. NIHR Great Ormond Street Hospital Biomedical Research Centre, University College London
   Overview
   Methods
   Assay n Number of
   Samples
   Number of
   Donors
   10x Visium 113,927 spots 30 10
   snRNA-seq 77,604 nuclei 19 10
   !
   Spatially resolved transcriptomics enables analysis
   of molecular organization of the human DLPFC
   !
   Unbiased spatial domains, identified with 10x
   Visium data and un-supervised clustering, look
   beyond classic histological layers
   !
   Integration with single nucleus RNA-sequencing
   data revealed distinct cell type compositions and
   cell-cell interactions within spatial domains
   !
   Enrichment analysis link genes associated with
   neuropsychiatric disorders to discrete spatial
   domains.
   !
   The research results and integrated datasets are
   available to the scientific community at
   research.libd.org/spatialDLPFC/
   Access this Dataset
   Spatial Data Shiny App
   snRNA-seq on iSEE App
   preprint SpatialDLPFC
   Interactive
   Websites
   SpatialLIBD
   R Package &
   Data
   Download
   Poster
   speakerdeck.com/
   lahuuki
   Acknowledgements
   Maynard et al., Nat Neurosci, 2021, 10.1038/s41593-020-00787-0
   Zhao et al., Nat BioTech, 2021, 10.1038/s41587-021-00935-2
   Emani et al, 2022, (syn30106435) 10.7303/syn4921369
   Presenter
   Louise
   Huuki-Myers
   Staff Scientist at LIBD
   @lahuuki
   lahuuki.github.io
   PsychENCODE Consortium
   snRNA-seq
   Identification of Data-Driven Spatial Domains
   BayesSpace: spatially-aware unsupervised clustering Spk
   Ds
   • K = 2: white matter vs. grey matter
   • K = 9: classic histological layers
   • K = 16: laminar with 2+ domains per histological layer
   k = number of domains
   s = domain number
   DE Genes in Spatial Domains
   Sp9
   D1
   is enriched for CLDN5: vascular domain (endothelial cells)
   Pairwise DE for Layer 1 sub-domains
   1
   2
   logcounts
   min > 0
   CLDN5
   1
   2
   3
   logcounts
   min > 0
   1
   2
   3
   logcounts
   min > 0
   Identified 29 fine resolution clusters across 70k nuclei
   Annotation
   Confidence
   X - high
   * - low
   Spot Deconvolution
   Spatially Map Disease Ligand Receptor Interactions
   Spatial Registration of Neuropsychiatric Data Sets
   • Benchmarked 3 spot deconvolution methods with Visium-SPG
   • Predicted cell type composition of spatial domains with
   Cell2Location & Tangram – produced different results
   • Identified Schizophrenia associated LR pairs in cell-cell
   communication analysis
   • Found co-localization of EFNA5 & EPHA5 in Sp9
   D7
   ~L6
   • Demonstrate reproducibility of snRNA-seq cell type population
   spatial registration
   • Gene set enrichment analysis of DE genes from ASD and PTSD
   data show enrichment with specific spatial domians
   Annotate 12 layer specific populations
   Paired Single Nucleus RNA-seq
   Tutorial: Spatial
   Registration
   

   View Slide