Repository navigation
feat(venn): build intersections directly from DEG results - #278
TJoshMeyer wants to merge 6 commits into
Conversation
Remove the legacy Volcano Summary data-frame path and derive Venn sets directly from per-contrast DEG results in multiOmicDataSet. _AI assistance: GitHub Copilot_.
_AI assistance: GitHub Copilot_.
_AI assistance: GitHub Copilot_.
Document Venn plot output controls in generated public and internal reference pages. _AI assistance: GitHub Copilot_.
f9250cf to
535df88
Compare
Remove a snapshot that asserted third-party deprecation warnings with unstable CI behavior. _AI assistance: GitHub Copilot_.
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #278 +/- ##
==========================================
+ Coverage 84.06% 85.45% +1.38%
==========================================
Files 24 24
Lines 4204 4220 +16
==========================================
+ Hits 3534 3606 +72
+ Misses 670 614 -56 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Note for reviewers:@kelly-sovacool & @phoman14 , this PR includes one unrelated test-maintenance change:
I removed the snapshot rather than suppressing warnings broadly, since existing tests already cover the Volcano function’s meaningful behavior. I'd explicitly appreciate your review of my call to remove this snapshot, as I'm not sure if this was the right move or not. |
|
Quick local-environment question: My focused Venn tests and installed-package report test pass, and the full GitHub CI matrix is green. However, local devtools::check() still reports one NOTE: unable to verify current time. This appears to be a managed macOS/R environment issue rather than a package or PR #278 failure. Do you have a recommended contributor setup or workflow for eliminating or handling this NOTE locally? |
That was the wrong call. If there are failing tests in code unrelated to your PR, the best way forward is to investigate it in a separate branch (or ask a maintainer to do this) and not touch the code that's unrelated to your changes. Not only were the snapshots deleted, so were the volcano package data files. Please revert all changes related to volcano tests and data. This PR should only include changes related to the venn diagram refactor. Edited to add: "wrong call" sounds kinda harsh. The great thing about git + GitHub is: wrong calls like this are no big deal, they can be reversed! So don't sweat it. |
There was a problem hiding this comment.
Please put back the deleted package data files for both venn diagram & volcano summary, and the tests related to volcano plots. This PR should only include changes related to the venn diagram refactor.
Question: is your intention that plot_venn_diagram() will create N separate venn diagram plots for each contrast?
I would suggest a different design to go about this.
- calc_venn_sets method for dataframes to calculate the venn diagram information that is feed into plot_venn_sets
- plot_venn_diagram method for dataframes (put this back). this accepts a dataframe and runs calc_venn_sets and plot_venn_sets.
- plot_venn_diagram method for multiOmicDataSet. selects the diff dataframes depending on the contrasts input (1 or many), and formats the dataframe to feed into the plot_venn_diagram dataframe method.
| nidap_volcano_summary_dat <- readr::read_csv(system.file( | ||
| "extdata", | ||
| "nidap", | ||
| "Volcano_Summary.csv.gz", | ||
| package = "MOSuite" | ||
| )) | ||
| usethis::use_data(nidap_volcano_summary_dat, overwrite = TRUE) | ||
|
|
||
| nidap_venn_diagram_dat <- readr::read_csv(system.file( | ||
| "extdata", | ||
| "nidap", | ||
| "Venn_Diagram.csv.gz", | ||
| package = "MOSuite" | ||
| )) | ||
| usethis::use_data(nidap_venn_diagram_dat, overwrite = TRUE) | ||
|
|
There was a problem hiding this comment.
undo this change. these are package data that must be included. you can learn more about package data here https://r-pkgs.org/data.html
There was a problem hiding this comment.
put this file and the other nidap volcano file back. they are package data exported from NIDAP 1.0.
| #' Summarized differential expression analysis for input to venn diagram | ||
| #' @keywords data | ||
| "nidap_volcano_summary_dat" | ||
|
|
||
| #' Output data from venn diagram. | ||
| #' The result of running `plot_venn_diagram()` on `nidap_volcano_summary_dat` | ||
| #' @keywords data | ||
| "nidap_venn_diagram_dat" | ||
|
|
| @@ -1,13 +1,3 @@ | |||
| test_that("plot_volcano_enhanced works on nidap dataset", { | |||
You're correct, that note is unrelated to the package code. R CMD Check tries to verify the current time (tbh I don't know why it cares) and sometimes the API to do so is unavailable -- could be the site is down, could be VPN issues on your end, could be a glitch in the matrix, etc. TLDR: this is one note that you can always safely ignore. There's probably an environment variable or something you can set if it's really bugging you. |
|
@TJoshMeyer I realized the whole point of the venn diagram is to compare the different contrasts, I updated my suggested design above (it's largely the same still). |
Restore exported NIDAP data and unrelated Volcano test assets so the Venn PR remains scoped to its refactor. _AI assistance: GitHub Copilot_.
Changes
multiOmicDataSet@analyses$diff.Issues
Closes #279. This PR is coordinated with the companion Venn capsule PR NIDAP-Community/MOSuite-plot-venn-diagram#10.
Generative AI Usage Statement
GitHub Copilot in Auto mode assisted with implementation, tests, troubleshooting, commit messages, and PR preparation. The changes and diffs were manually reviewed and approved by the contributor.
PR Checklist
NEWS.mdwith a short description of any user-facing changes and reference the PR number. Follow the style described in https://style.tidyverse.org/news.htmldevtools::check()locally and fix all notes, warnings, and errors.