2023-NIA-LNG-Seminar-Series

Transcript

1. @lcolladotor
   lcolladotor.github.io
   lcolladotor.github.io/bioc_team_ds
   Harnessing the power of
   spatially-resolved transcriptomics
   one step at a time (part 2/2)
   Leonardo Collado Torres, Investigator #Visium_SPG_AD
   2023 NIA/LNG Neurodegenerative Diseases Seminar Series
   June 09 2023
   Slides available at speakerdeck.com/lcolladotor
   

   View Slide

2. twitter.com/sanghokwon17/status/1650589385379962881 from 2023-04-24
   Sang Ho Kwon
   @sanghokwon17
   DOI: 10.1101/2023.04.20.537710
   #Visium_SPG_AD
   In collaboration with
   Download data with
   spatialLIBD::fetch_data()
   research.libd.org
   /Visium_SPG_AD
   

   View Slide

3. Visium spatial clustering works for variables with high %
   variance explained. But what about other ones?
   DOI: 10.1038/s41593-020-00787-0
   

   View Slide

4. Alzheimer’s disease related neuropathology
   Adopted and modified from B Wang (2018) and the Brain from the Top to Bottom in McGill University
   20 um
   Sang Ho Kwon
   Amyloid beta
   pTau
   

   View Slide

5. Hypothesis
   Local tissue microenvironments in close proximity to AD-related
   neuropathology have distinct cellular and molecular signatures. Sang Ho Kwon
   

   View Slide

6. Visium Spatial Proteogenomics (Visium-SPG)
   Visium-SPG = Visium SRT + immunofluorescence
   (using identical tissue samples)
   Sang Ho Kwon
   

   View Slide

7. Experimental design & study overview
   Braak V-VI & CERAD frequent
   Sang Ho Kwon
   

   View Slide

8. Donor demographics
   BrNum
   Age of
   death
   (years
   old)
   Sex Ancestry
   PMI
   (hours)
   Dx
   Best RIN
   PFC
   Braak CERAD APOE
   # Visium
   Replicates
   Br3854 65.75 Female European 31.5 Alzheimer 7.0 Stage VI Frequent e3/e4 2
   Br3873 88.78 Female European 29.0 Alzheimer 7.2 Stage V Frequent e3/e3 2
   Br3880 90.47 Male European 35.0 Alzheimer 7.1 Stage VI Frequent e3/e3 3
   Br3874 73.05 Male European 13.5 Control 7.2 Stage IV None e2/e3 3
   DOI: 10.1101/2023.04.20.537710
   

   View Slide

9. AD pathology signal is too small to detect by
   spatially-resolved gene expression alone research.libd.org/Visium_SPG_AD/
   

   View Slide

10. Visium
    * ~5k spots in honeycomb
    * gene expression per spot
    * tissue (H&E staining)
    Immunofluorescence (IF)
    * multi-channel (6) images
    * identifies morphological features of interest
    * large: might be broken in tiles
    Channel 1
    * triangle feature
    Channel 2
    * cloud feature
    Channel 6
    * xyz feature
    Tissue (bright field image)
    Visium spot
    Channel 1 feature
    Channel 2 feature
    +
    Visium Spatial Proteogenomics (Visium-SPG / IF) raw data: 2 types
    

    View Slide

11. Spot ID Gene 1 Gene 2 Gene X
    spot0001 0 12 39
    spot0002 4 0 27
    Spot ID Gene 1 Gene 2 Gene X In Tissue # cells
    spot0001 0 12 39 true 3
    spot0002 4 0 27 false 0
    * spaceranger
    * Loupe Browser
    * VistoSeg on H&E bright
    field image
    Visium Analysis
    @MadhaviTippani
    @HeenaDivecha
    cell
    VistoSeg DOI:
    10.1101/2021.08.04.452489
    

    View Slide

12. Feature ID X center Y center type intensity
    feat0001 5 102 triangle 130.4
    feat0002 10 30 cloud 99.1
    Max (X, Y)
    Min (X, Y)
    Area
    ...
    SPG / IF Image Analysis
    * segment each
    channel
    * find features
    Challenges:
    * morphological
    features can be
    quite diverse
    * images are large
    * multiple tiles
    +
    

    View Slide

13. Spot ID # Triangle # Cloud % triangle % cloud
    spot0001 0 12 0 17
    spot0002 4 0 27 0
    Merge Visium & IF
    IF
    Spot ID Gene 1 Gene 2 Gene X In Tissue # cells
    spot0001 0 12 39 true 3
    spot0002 4 0 27 false 0
    Visium
    downstream
    * QC
    * analyses
    spatialLIBD DOI:
    10.1186/s12864-022-08601-w
    VistoSeg DOI:
    10.1101/2021.08.04.452489
    Check version 2!!
    @MadhaviTippani
    

    View Slide

14. Estimating pathological burden per spot
    to generate transcriptome-scale maps of AD pathology
    1. both >
    2. A𝛽 or pTau >
    3. next_both >
    4. next_A𝛽 or next_pTau >
    5. none
    

    View Slide

15. AD pathology signal is too small to detect by
    spatially-resolved gene expression alone research.libd.org/Visium_SPG_AD/
    

    View Slide

16. Pathology burden is more common in the gray matter
    BayesSpace with 2 clusters:
    Cluster 1: gray matter (GM)
    Cluster 2: white matter (WM)
    1
    2
    

    View Slide

17. Identifying transcriptional signatures of AD-related neuropathology
    Sang Ho Kwon
    

    View Slide

18. Identifying transcriptional signatures of Abeta-related neuropathology
    Sang Ho Kwon
    

    View Slide

19. Validation study design using RNAscope FISH-IF
    Sang Ho Kwon
    

    View Slide

20. HALO-based image analysis
    

    View Slide

21. 6 Abeta-associated DEGs on RNAscope FISH-IF
    (Kruskal-Wallis test,
    *p<0.05, &p<0.005, and #p<0.0001)
    Sang Ho Kwon
    

    View Slide

22. Visium + RNAscope
    to provide additional insights into spatial gene expression gradients
    Sang Ho Kwon
    

    View Slide

23. Identifying transcriptional signatures of Abeta microenvironment
    Previously not significant Sang Ho Kwon
    A𝛽 + next_A𝛽 = A𝛽_env (environment)
    

    View Slide

24. bioconductor.org/packages/spatialLIBD
    Pardo et al, 2022 DOI 10.1186/s12864-022-08601-w
    Maynard, Collado-Torres, 2021 DOI 10.1038/s41593-020-00787-0
    Brenda Pardo Abby Spangler
    @PardoBree @abspangler
    Louise A. Huuki-Myers
    @lahuuki
    

    View Slide

25. Conclusions
    • Visium-SPG can identify local transcriptional signatures of late-stage AD pathology
    • RNAscope FISH-IF can further explore spatial gene expression patterns of a few select
    features
    • Proof-of-concept study: our work lays the groundwork. Can now use Visium-SPG for other
    diseases too
    • Future studies are needed to produce mechanistic insights
    • Explore our code & data at research.libd.org/Visium_SPG_AD & download it with
    spatialLIBD::fetch_data()
    ○ There’s lots more to explore with this data!
    Sang Ho Kwon
    Slide adapted from:
    

    View Slide

26. twitter.com/sanghokwon17/status/1650589385379962881 from 2023-04-24
    Sang Ho Kwon
    @sanghokwon17
    DOI: 10.1101/2023.04.20.537710
    #Visium_SPG_AD
    In collaboration with
    Download data with
    spatialLIBD::fetch_data()
    research.libd.org
    /Visium_SPG_AD
    

    View Slide

27. @MadhaviTippani
    Madhavi Tippani
    @HeenaDivecha
    Heena R Divecha
    @lmwebr
    Lukas M Weber
    @stephaniehicks
    Stephanie C Hicks
    @abspangler
    Abby Spangler
    @martinowk
    Keri Martinowich
    @CerceoPage
    Stephanie C Page
    @kr_maynard
    Kristen R Maynard
    @lcolladotor
    Leonardo Collado-Torres
    @Nick-Eagles (GH)
    Nicholas J Eagles
    Kelsey D Montgomery
    Sang Ho Kwon
    Image Analysis
    Expression Analysis
    Data Generation
    Thomas M Hyde
    @lahuuki
    Louise A Huuki-Myers
    @BoyiGuo
    Boyi Guo
    @mattntran
    Matthew N Tran
    @sowmyapartybun
    Sowmya Parthiban
    Slides available at
    speakerdeck.com
    /lcolladotor
    + Many more LIBD, JHU, and
    external collaborators
    @mgrantpeters
    Melissa Grant-Peters
    @prashanthi-ravichandran (GH)
    Prashanthi Ravichandran
    

    View Slide

28. #GBD23 Thank you for having us over in the UK 󰏅!
    !Gracias Pilar & Cornelis for the invitation to present at NIA/LNG!
    

    View Slide

29. lcolladotor.github.io
    @lcolladotor
    

    View Slide