Fix the two sided test on IBMA Fishers and Stouffers estimators

[JulioAPeraza]

Closes None.

Changes proposed in this pull request:

• As noticed in PyMARE, combination tests only take z-scores from one-sided p-values. Before, we were performing a concordant test, but that returned the values of the left tail to positive as well.
• To be able to handle both tails, we perform the test for each tail separately and take the values with minimum p-value as output.

Summary by Sourcery

Fix the two-sided test implementation for IBMA Fishers and Stouffers estimators by running separate tests for each tail and using the minimum p-value to ensure correct handling of both tails.

Bug Fixes:

• Fix the two-sided test for IBMA Fishers and Stouffers estimators by separately handling each tail and selecting the minimum p-value.

[sourcery-ai]

Reviewer's Guide by Sourcery

This pull request fixes the implementation of two-sided tests for Fisher's and Stouffer's estimators in the IBMA (Image-Based Meta-Analysis) module. The changes modify how z-scores and p-values are calculated to properly handle both positive and negative tails in two-sided tests.

File-Level Changes

Change	Details	Files
Implement separate calculations for positive and negative tails in two-sided tests	Remove the '_mode' attribute from both Fishers and Stouffers classes Add separate estimators for positive and negative tails Implement logic to choose the minimum p-value between positive and negative tails Create z-map based on which tail has the smaller p-value Update the return values to use the new calculation method	nimare/meta/ibma.py
Modify Fisher's method to handle two-sided tests correctly	Add conditional logic for two-sided tests Create separate PyMARE datasets for positive and negative tails Implement minimum p-value selection for two-sided tests Update z-map calculation based on selected p-values	nimare/meta/ibma.py
Update Stouffer's method to properly handle two-sided tests	Remove single estimator approach and implement separate estimators for each tail Add conditional logic for two-sided tests Create separate PyMARE datasets for positive and negative tails Implement minimum p-value selection for two-sided tests Update z-map calculation based on selected p-values	nimare/meta/ibma.py

---

Tips

• Trigger a new Sourcery review by commenting @sourcery-ai review on the pull request.
• Continue your discussion with Sourcery by replying directly to review comments.
• You can change your review settings at any time by accessing your dashboard:
  • Enable or disable the Sourcery-generated pull request summary or reviewer's guide;
  • Change the review language;
• You can always contact us if you have any questions or feedback.

[sourcery-ai]

Hey @JulioAPeraza - I've reviewed your changes - here's some feedback:

Overall Comments:

• Consider adding explanatory comments for the two-sided test implementation to improve code readability and maintainability.
• To reduce code duplication, consider extracting the common logic for positive and negative tails into a separate function.

Here's what I looked at during the review

• 🟡 General issues: 1 issue found
• 🟢 Security: all looks good
• 🟢 Testing: all looks good
• 🟢 Complexity: all looks good
• 🟢 Documentation: all looks good

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment to tell me if it was helpful.}

[sourcery-ai]

suggestion (performance): Consider performance implications of running two separate tests for two-sided cases

The new implementation runs two separate tests when two_sided is True, which could potentially double the computation time. Consider if there's a way to optimize this, especially for large datasets.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 88.25%. Comparing base (2c9b52a) to head (23865c5).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #903      +/-   ##
==========================================
+ Coverage   88.23%   88.25%   +0.02%     
==========================================
  Files          48       48              
  Lines        6382     6394      +12     
==========================================
+ Hits         5631     5643      +12     
  Misses        751      751              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[JulioAPeraza]

@jdkent I checked the examples gallery, and the results look good now: https://nimare--903.org.readthedocs.build/en/903/auto_examples/02_meta-analyses/02_plot_ibma.html#sphx-glr-auto-examples-02-meta-analyses-02-plot-ibma-py

[jdkent]

I'm making a pull request here to fix the underlying bug in pymare: neurostuff/PyMARE#125

[JulioAPeraza]

Thanks! Closing this in favor of #904.