Use left join for DE Spearman/AUC (Fixes #227)

[beabevi]

Summary

• Compute DE Spearman LFC on real significant genes by left-joining pred DE and filling missing pred fold-change with 0.0.
• Compute PR/ROC AUC on all real-labeled genes by left-joining pred scores, filling missing pred_fdr with 1.0, and clipping to [1e-10, 1.0] before -log10.

Why

Current inner joins restrict metrics to the intersection of pred/real DE tables, which contradicts intended semantics. This PR aligns metrics with the real-DE / all-genes definitions.

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request refines the computation of differential expression (DE) Spearman and AUC metrics. By transitioning from inner joins to left joins and implementing robust handling for missing predicted values, the changes ensure that these metrics accurately reflect the intended semantics, considering all relevant real-labeled or significant genes rather than just the intersection of predicted and real data.

Highlights

• Spearman LFC Calculation: Modified the Spearman LFC calculation to use a left join instead of an inner join when merging real significant genes with predicted data. Missing predicted fold-change values are now filled with 0.0, ensuring that all real significant genes are considered.
• AUC Calculation: Updated the AUC calculation to use a left join when merging real-labeled genes with predicted data. Missing predicted FDR values are now filled with 1.0 and clipped to the range [1e-10, 1.0] before applying the -log10 transformation, aligning the metric with the definition for all real-labeled genes.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• src/cell_eval/metrics/_de.py
  • Changed the join type from 'inner' to 'left' for Spearman LFC calculation.
  • Added a step to fill null predicted fold-change values with 0.0 in Spearman LFC.
  • Changed the join type from 'inner' to 'left' for AUC calculation.
  • Added logic to fill null predicted FDR values with 1.0 and clip them to [1e-10, 1.0] before log transformation in AUC calculation.

Activity

• No human activity has been recorded on this pull request yet.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request correctly changes the join strategy from inner to left for calculating Spearman correlation and AUC metrics. This ensures that all relevant genes from the ground truth data are considered, providing more accurate metric calculations. The changes in DESpearmanLFC and compute_generic_auc are logical and well-implemented. I have one minor suggestion to improve code clarity in compute_generic_auc.

[gemini-code-assist]

The update to pred_fdr_col via pred_q.alias(pred_fdr_col) is not necessary, as this column isn't used again within the function. You can simplify this with_columns call by only creating the nlp column.

        .with_columns(
            (-pred_q.log10()).alias("nlp")
        )