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.

Sign up for GitHub

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

Jump to bottom

Correlated pvalue #9

Closed
tleonardi opened this issue Oct 10, 2018 · 2 comments
Closed

Correlated pvalue #9

tleonardi opened this issue Oct 10, 2018 · 2 comments
Labels
TODO Something that needs to be implemented

Comments

@tleonardi
Copy link
Owner

tleonardi commented Oct 10, 2018 •
edited

The pvalues for neighbouring positions are highly correlated, thus violating the assumptions of Fisher's method for combining p-values.
Based on the benchmarking done in this paper, Hou's method seems the best solution.
@tleonardi tleonardi added the TODO Something that needs to be implemented label Oct 10, 2018
@a-slide
Copy link
Collaborator

a-slide commented Oct 10, 2018

That is absolutely True. Another option, at least for the kmean strategy, is a multidimentional kmean combining all adjacent positions instead of position per position and then combining pvalues.

tleonardi pushed a commit that referenced this issue Oct 11, 2018
Implemented Hou's method, issue #9
413bb33
@tleonardi tleonardi mentioned this issue Oct 11, 2018
Closed
@tleonardi
Copy link
Owner Author

Discussion about multidimensional kmean moved to #10

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
TODO Something that needs to be implemented
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@a-slide @tleonardi