add curation filtering on Shiny for network interpretation

[tonywu1999]

Pending: Removal of API key

Summary by CodeRabbit

• New Features
  • Added a "Filter out statements curated as incorrect" checkbox in the Data Upload section to exclude curated-incorrect statements from network generation, producing cleaner, more reliable visualizations.
  • Enhanced UI tooltips (including the Force Include Proteins tooltip) to clarify search behavior and available search options.
  • Checkbox defaults to off when unspecified to ensure consistent behavior.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

📝 Walkthrough

Walkthrough

Adds a boolean filterByCuration option (UI checkbox) to the visualize-network module; the value is collected by getInputParameters, propagated through extractSubnetwork/visualizeNetworkServer, and passed as filter_by_curation to getSubnetworkFromIndra and generated network code.

Changes

Cohort / File(s)	Change Summary
Curation Filter UI Component R/module-visualize-network-ui.R	Added exported createCurationFilterCheckbox(ns) to render a "Filter out statements curated as incorrect" checkbox with tooltip; integrated into the Data Upload UI.
Server-side Parameter Propagation & Codegen R/module-visualize-network-server.R	getInputParameters now includes filterByCuration (defaults NULL→FALSE). extractSubnetwork(…) signature gains filterByCuration and forwards it as filter_by_curation (with api_key = "") to getSubnetworkFromIndra. visualizeNetworkServer and generate_network_code updated to pass/emit filter_by_curation = params$filterByCuration.

Sequence Diagram(s)

sequenceDiagram
    participant UI as UI (checkbox)
    participant Server as visualizeNetworkServer
    participant Params as getInputParameters
    participant Extract as extractSubnetwork
    participant Indra as getSubnetworkFromIndra
    participant Codegen as generate_network_code

    UI->>Server: user toggles filterByCuration
    Server->>Params: collect input params (includes filterByCuration)
    Params-->>Server: params{filterByCuration}
    Server->>Extract: call extractSubnetwork(params, filterByCuration)
    Note right of Extract: filterByCuration forwarded\nas filter_by_curation
    Extract->>Indra: getSubnetworkFromIndra(..., filter_by_curation=BOOL, api_key="")
    Server->>Codegen: generate_network_code(params) (emits filter_by_curation=params$filterByCuration)
    Codegen-->>Server: generated network construction code
    Server-->>UI: render updated network

Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• feature(network-interpretation): Add feature to enable user to search for any identifier in subnetwork #120 — Modifies network visualization/server wiring to pass additional user-controlled parameters into getSubnetworkFromIndra (closely related propagation changes).
• Add INDRA integration to MSstatsShiny #110 — Introduced/extended the network module parameter (filterByCuration) propagation into extractSubnetwork/codegen; directly connected to the same functions/files.

Poem

🐰 I found a tiny checkbox in the meadow of code,
"Filter wrong statements" — I nudged it down the road.
I carried that boolean through server and stream,
Hopped past indra and codegen — a tidy little dream.
🥕

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'add curation filtering on Shiny for network interpretation' directly and clearly describes the main change: adding curation filtering functionality to the Shiny UI for network visualization.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[github-actions]

Failed to generate code suggestions for PR

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In @R/module-visualize-network-server.R:
- Around line 616-618: The generated code now adds filter_by_curation but the
server runtime path forces api_key = "" creating a mismatch; update
generate_network_code to include api_key = "" when the server path forces it, or
better remove the forced api_key = "" in the server runtime so both
code-generation (generate_network_code) and server runtime rely on the same
default behavior; locate generate_network_code and the server runtime block that
sets api_key = "" and either add the same forced api_key parameter to
generate_network_code or eliminate the hard-coded api_key = "" in the server
path so both paths are consistent.

🧹 Nitpick comments (1)

R/module-visualize-network-server.R (1)

244-260: Prefer isTRUE() to coerce filterByCuration into a strict scalar boolean.

Current code works, but as.logical() can yield NA if the input ever becomes a non-boolean string; isTRUE() safely normalizes NULL/FALSE/TRUE into FALSE/TRUE.

Proposed refactor

-  filterByCuration <- if(is.null(input$filterByCuration)) {
-    FALSE  # Default to FALSE if somehow NULL 
-  } else {
-    as.logical(input$filterByCuration)
-  }
+  filterByCuration <- isTRUE(input$filterByCuration)

📜 Review details

Configuration used: defaults

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between c4c6f7a and b1ff54d.

📒 Files selected for processing (1)

• R/module-visualize-network-server.R

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: build

🔇 Additional comments (2)

R/module-visualize-network-server.R (2)

531-535: Propagation of filterByCuration into extractSubnetwork() looks consistent.

---

286-303: This review comment is based on incorrect assumptions about parameter support. Both filter_by_curation and api_key are documented, supported parameters in getSubnetworkFromIndra() and will not cause runtime errors. Passing api_key = "" is equivalent to the function default, so no change is necessary.

Likely an incorrect or invalid review comment.

[coderabbitai]

⚠️ Potential issue | 🟡 Minor

Generated code now includes filter_by_curation, but runtime path also forces api_key = "" (mismatch).

If you keep api_key = "" in the server path, consider reflecting the same behavior in generate_network_code (or, preferably, remove the forced empty key in the server path so both paths rely on the same default behavior).

🤖 Prompt for AI Agents

In @R/module-visualize-network-server.R around lines 616 - 618, The generated
code now adds filter_by_curation but the server runtime path forces api_key = ""
creating a mismatch; update generate_network_code to include api_key = "" when
the server path forces it, or better remove the forced api_key = "" in the
server runtime so both code-generation (generate_network_code) and server
runtime rely on the same default behavior; locate generate_network_code and the
server runtime block that sets api_key = "" and either add the same forced
api_key parameter to generate_network_code or eliminate the hard-coded api_key =
"" in the server path so both paths are consistent.