Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Warning for ties in stats #20

Merged
merged 3 commits into from
Feb 21, 2018
Merged

Warning for ties in stats #20

merged 3 commits into from
Feb 21, 2018

Conversation

ToledoEM
Copy link
Contributor

Print a message if there are ties in the metric and the percentage of it.

@assaron,

Due to the previous discussion about ties (#18) and never ending analysis (#19), I put some changes to include a warning for the presence of ties in the stats:

  • This only for metric not equal to zero, as changing the order of genes equal to 0 it is no relevant.
  • I do not know which percentage of ties are aceptable, this is still open.
  • Ties shouldn't impact in the biological interpretation of the results. But the user should be aware of this when they are doing the analysis. As discussed in Preranked metric ties #18 and fgsea never ending #19.

Best,
ToledoEM

Print a message if there are ties in the metric and the percentage of it.    
This only for metric not equal to zero, as random order of genes equal to 0 it is no relevant.    

No idea which percentage of ties are aceptable, nor if it is relevant for the biological interpretation of the results. But the user should be aware of this when they are doing the analysis. As discussed in #18 and #19
@codecov-io
Copy link

codecov-io commented Feb 16, 2018

Codecov Report

Merging #20 into master will increase coverage by 0.02%.
The diff coverage is 100%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master      #20      +/-   ##
==========================================
+ Coverage   93.52%   93.54%   +0.02%     
==========================================
  Files           5        5              
  Lines         710      713       +3     
==========================================
+ Hits          664      667       +3     
  Misses         46       46
Impacted Files Coverage Δ
R/fgsea.R 85.47% <100%> (+0.14%) ⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 5d5218b...1646d12. Read the comment docs.

@assaron
Copy link
Member

assaron commented Feb 19, 2018

Good point about ties for zero ranks!

Could you add a test for these cases?

@ToledoEM
Copy link
Contributor Author

Sorry, I do not get what you mean with a test for the zero rank cases.

@assaron
Copy link
Member

assaron commented Feb 20, 2018

I mean, adding testthat tests somewhere here https://github.com/ctlab/fgsea/blob/master/tests/testthat/test_gsea_analysis.R so that code is covered by tests.

Testthat test for  "Ties detection in ranking works"
@assaron assaron merged commit 1646d12 into ctlab:master Feb 21, 2018
@assaron
Copy link
Member

assaron commented Feb 21, 2018

Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants