- Each student will submit the activity individually to Canvas, but students are encouraged to help one another complete the work
- Complete the Statistics of Gene Expression Activity found in the DataComputing eBook
- make sure you carefully follow the instructions and mirror the code in the activity as you work
- every task in the activity should have narrative text describing your observations; most steps also require code chunks and corresponding output.
- you should make at least 3 commits in GitHub as you complete the activity
- submit the completed R Notebook to Canvas before deadline
Assignment is worth a total of 15 points.
- [5 points] Follow along with the activity in the book and reproduce the following figures as they appear in the book
- Figure 18.1: A bar chart comparing TOP3A expression in the different tissue types.
- Figure 18.2: TOP3A expression in the individual cells.
- Figure 18.3: TOP3A Bar chart with individual cells overlaid
- Figure 18.4: TOP3A Dynamite plot (bar chart with error bars)
- Figure 18.5: "Better than Dynamite" (TOP3A expression in the individual cells with overlaid confidence intervals by tissue type) - [1 point] Figure 18.1 Critique
- [2 point] Your turn: Pick your own probe and make a figure like that of Figure 18.5.
- [2 points] recreate Figure 18.6 showing 30 Probes with largest r-squared for expression level explained by tissue type
- [2 points] Your Turn Choose one probe with high R-squared. Plot expression versus tissue type (like Figure 18.5)
- [2 points] recreate Figure 18.7 (overlaid density plots)
- [1 points] recreate Figure 18.8 (comparison of highest R^2 from actual data and the Null)
- [1 point] Push changes to GitHub
In the "Probing for a Probe" section, you're introduced to the dplyr::do( ) function... this step is "slow" (i.e. takes a long time). Your code will actually fit a regression model to each of more than 32,000 probes and then store the R-squared for each one of them. In fact, if you understand what it's doing, it happens incredibly fast... but it still takes a long time (i.e., a couple minutes depending on your available computational resources).
- For the critique of Figure 18.1, make sure to address what about the data is not being displayed in Figure 18.1.
- AFTER you have written your critique of Figure 18.1, among the "several suggestions for improving the graphic", #3 and #4 will be most handy
- You'll need to combine elements of previous plots in order to produce Figure 18.5
- When you reproduce Figure 18.5 for a new Probe in the Your Turn portion, you shouldn't start from scratch. Find code earlier in the assignment that you can copy, paste, and modify to investigate the Probe of interest.
- ... Gene Expression: https://en.wikipedia.org/wiki/Gene_expression
- ...
$R^2$ : https://en.wikipedia.org/wiki/Coefficient_of_determination - ... False Discovery Rates: https://en.wikipedia.org/wiki/False_discovery_rate
- ... Approaches to simulating null hypothesis: https://bookdown.org/ajsage/statistics_for_data_science_notes/simulation-based-hypothesis-tests.html
- Side note: this is what my current research is about. If you're interested, I can tell you all about it!