RNAseq: A five course meal

These slides are from a talk I gave as part of a two-week next-generation sequencing analysis workshop (https://angus.readthedocs.io/en/2019/). In this talk, I provide a framework for understanding RNA-seq workflows that is based on the R for Data Science workflow of "import, tidy, transform, visualize, model, and communicate" mantra (https://r4ds.had.co.nz/). I also provide strategies for "healthy" workflows by making analogies to a five-course meal.

Transcript

1. RNAseq:
   A five course meal
   #DIBSI2019
   Rayna M. Harris
   @raynamharris
   1
   

   View Slide

2. Rayna M. Harris
   Postdoctoral scholar at UC Davis
   Scientist, Educator, Community Builder, Translator
   2
   

   View Slide

3. Special thanks
   to the
   Data Intensive Biology Lab
   and the
   Birds, Brain, & Banter Lab
   http://calisilab.ucdavis.edu/
   http://ivory.idyll.org/lab/
   3
   

   View Slide

4. I learned R and RNAseq in communities of practice
   The University of Texas at Austin, University California, Davis, Data Carpentry, Software Carpentry,
   The Carpentries-es, @cienciaPR @RLadiesGlobal @RLadiesBA @r4ds_es
   #DISBI2018
   4
   

   View Slide

5. Who reads books?
   5
   

   View Slide

6. Read R for Data Science
   6
   

   View Slide

7. R 4 Data Science workflow
   https://r4ds.had.co.nz/ 7
   

   View Slide

8. Data science as a five course meal
   8
   

   View Slide

9. Data snacks prevent “hanger”
   Appetizer
   9
   

   View Slide

10. Data snacks and source code
    `library()`
    `source()`
    `data()`
    Substantial, complex, noteworthy, valuable
    Provides insights into what follows
    10
    

    View Slide

11. Soup & salad, tidy then transform
    Salad Soup
    11
    

    View Slide

12. Data wrangling
    Appetizer
    Salad Soup
    Wrangle
    12
    

    View Slide

13. Data wrangling
    Simultaneously fun and painful
    Learn from your mistakes
    Write good documentation
    Write tests
    13
    

    View Slide

14. The tidyverse and scientific python are
    your data wrangling friends
    14
    

    View Slide

15. The DESeq2 salad
    http://bioconductor.org/packages/devel/bioc/vignettes/DESeq2/inst/doc/DESeq2.html 15
    

    View Slide

16. DESeq2 colData and counts
    https://biojupies.cloud/notebook/AHnOIzJEq 16
    

    View Slide

17. ncol(counts) == nrow(colData)
    https://biojupies.cloud/notebook/AHnOIzJEq 17
    

    View Slide

18. Summarized experiment soup
    http://bioconductor.org/packages/devel/bioc/vignettes/DESeq2/inst/doc/DESeq2.html#differential-expression-analysis 18
    

    View Slide

19. “You can’t have any pudding if you don’t eat yer meat.
    How can you have any pudding if you don’t eat yer meat?”
    Pink Floyd
    Dessert
    19
    

    View Slide

20. “Communication breakdown, it’s always the same
    I’m having a nervous breakdown, drive me insane.”
    Led Zeppelin
    Nuts
    20
    

    View Slide

21. Procrastigraphing: excessive creation of data visualizations
    Communicate
    Nuts
    Dessert
    21
    

    View Slide

22. Modeling is challenging. So many choices.
    Main course
    22
    

    View Slide

23. If you were at a banquet, would you order?
    A. Lobster
    B. Steak
    C. Lobster and steak
    D. Other
    E. Lobster, steak, and other
    23
    

    View Slide

24. If you were an RNAseq workflow, would you use:
    A. R
    B. Python
    C. R and Python
    D. Other
    E. An R/Python/other mix
    24
    

    View Slide

25. FlavoRs of differential gene expression models
    library("DESeq2")
    dds <- DESeqDataSetFromMatrix(countData = cts,
    colData = coldata,
    design = ~ condition)
    results(dds, contrast=c("condition","B","A"))
    http://bioconductor.org/packages/devel/bioc/vignettes/DESeq2/inst/doc/DESeq2.html#differential-expression-analysis
    25
    

    View Slide

26. Strategies for a well-balanced RNAseq analysis
    R and/or Python or other
    library()
    data()
    counts dds
    colData
    Rmarkdown
    Jupyter
    GitHub pages
    ~ condition
    Wrangle
    Explore
    ggplot()
    matplotlib
    Communicate
    26
    

    View Slide

27. BioJupies for instant
    RNA-seq visualizations
    in the cloud
    https://biojupies.cloud/notebook/AHnOIzJEq
    https://twitter.com/i/status/1062745599538282498
    27
    

    View Slide

28. https://biojupies.cloud/notebook/AHnOIzJEq
    Quickly assess sample quality and variation
    28
    

    View Slide

29. https://biojupies.cloud/notebook/AHnOIzJEq
    Quickly view results of gene ontology enrichment analyses
    29
    

    View Slide

30. https://biojupies.cloud/notebook/AHnOIzJEq
    Quickly explore patterns of differential gene expression
    30
    

    View Slide

31. https://biojupies.cloud/notebook/AHnOIzJEq
    But how many dots are red and blue?
    ?
    ?
    31
    

    View Slide

32. https://biojupies.cloud/notebook/AHnOIzJEq
    But, what if I want to view many volcano plots?
    32
    

    View Slide

33. Convert .Rmd files
    to GitHub pages to
    communicate results
    https://macmanes-lab.github.io/austinCORT/
    33
    

    View Slide

34. Open source options for programming
    34
    

    View Slide

35. My progress: from a novice with data and tools
    https://github.com/raynamharris/DissociationTest
    Harris, Kao, Alarcón, Hofmann, Fenton 2017
    https://www.biorxiv.org/content/10.1101/153585v1
    35
    

    View Slide

36. To a practitioner with reproducible workflows
    https://github.com/raynamharris/DissociationTest
    Harris, Kao, Alarcón, Hofmann, Fenton 2019
    https://onlinelibrary.wiley.com/doi/10.1002/hipo.23095
    36
    

    View Slide

37. Onward to experiments with many factors
    https://github.com/macmanes-lab/DoveParentsRNAseq/
    Made for a poster at the Society for Behavioral Neuroendocrinology
    37
    

    View Slide

38. Identify time-like changes in specific genes
    https://github.com/macmanes-lab/DoveParentsRNAseq/
    38
    

    View Slide

39. Find meaningful principle components of variation
    https://github.com/macmanes-lab/DoveParentsRNAseq/
    39
    

    View Slide

40. Create functions to run all pairwise comparisons
    `contrast = c(“treatment”, “varB”, “varA”)`
    https://github.com/macmanes-lab/DoveParentsRNAseq/
    40
    

    View Slide

41. The risk of unintentional p-hacking
    http://www.stat.columbia.edu/~gelman/research/unpublished/p_hacking.pdf
    41
    

    View Slide

42. The risk of unintentional p-hacking
    http://www.stat.columbia.edu/~gelman/research/unpublished/p_hacking.pdf
    42
    

    View Slide

43. Exploratory analyses vs. model testing
    Explore
    43
    

    View Slide

44. 60% 20% 20%
    training set query set testing set
    Should we be analyzing all the data all the time?
    44
    

    View Slide

45. “Soup to nuts: setting up a new RNAseq analysis”Titus
    45
    

    View Slide

46. “A five course meal is an excellent analogy!”Rayna
    Appetizer Soup & Salad Dessert Nuts
    Main course
    :)
    46
    

    View Slide

47. Framework for understanding RNAseq workflows
    R and/or Python or other
    library()
    data()
    counts dds
    colData
    ggplot()
    matplotlib
    Rmarkdown
    Jupyter
    GitHub pages
    ~ condition
    47
    

    View Slide

48. Strategies for creating healthy RNAseq workflows
    R and/or Python or other
    library()
    data()
    counts dds
    colData
    ggplot()
    matplotlib
    Rmarkdown
    Jupyter
    GitHub pages
    ~ condition
    Communicate
    Wrangle
    Explore
    Model
    48
    

    View Slide

49. What questions
    do you have?
    @raynamharris
    49
    

    View Slide