[feature] Transition the phenotype associations to non-parametric tests #171
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
cdiener
force-pushed
the
feature/new_phenotype
branch
3 times, most recently
from
5f050a2
to
a4b2c7d
Compare
cdiener
force-pushed
the
feature/new_phenotype
branch
from
d2bcbe6
to
eb4eeb5
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Purpose
The former strategy of identifying metabolite:phenotype associations with coefficients from a LASSO models has proven to be quite unstable due to (a) the non-uniqueness of the coefficients and (b) instability across scikit-learn versions and initialization. This PR switches this to a more stringent approach that uses Mann-Whitney U or Spearman rho tests to assess metabolite-level associations. A LASSO models is still fit to assess the overall/global association with the phenotype.
Visualization
The visualization is similar but now shows a confusion matrix for binary outcomes. A quantitative effect measure will be used instead of coefficients.
example visualization
Side effects
This adds new example data sets to help test and document the new functionality.
TODO