[t0mdavid-m]

Requested by Tom · project thread

One PR for all open OpenDDA work, plus a pass over every page for a Biology PhD audience: plain-language explanations, sensible defaults, and a navigation that leads from upload to analysis with the important results up front.

Combined PRs

Merged unchanged (conflicts in app.py, results_filtered.py, results_helpers.py and WorkflowTest.py were import/page-list unions):

• Add an Identification QC page (score distribution, precursor error, PSMs per file) #42 Identification QC page
• Add a Proteins page: protein table with coverage, linked peptides and PSMs #43 Proteins page (coverage, linked peptides and PSMs)
• Filtered PSMs: fragment mass error plot and b/y ion table for the selected PSM #44 Fragment mass error plot and b/y ion table
• Search cRAP contaminants with the database, flag them and hide them from protein results #45 cRAP contaminants
• Add a Run Overview page: key figures and a per-file summary #46 Run overview
• Add a Samples page: PSMs, peptides and proteins per sample, by condition #47 Samples page
• PSM tables: show modifications as labels, ProForma or OpenMS; ProForma in exports #48 Modification display / ProForma
• Add an Open Search preset and a Mass Shifts page that matches shifts to Unimod #49 Open Search preset and Mass Shifts page
• Instrument presets: Orbitrap HCD, timsTOF and Ion Trap CID #50 Instrument presets
• Add PRIDE dataset import for reprocessing public data #51 PRIDE import
• deps: bump certifi from 2025.1.31 to 2026.1.4 #1–deps: bump markupsafe from 3.0.2 to 3.0.3 #5 dependabot bumps (certifi, idna, kiwisolver, markupsafe, urllib3)

Not included because main already has their content or they are stale against it: #38 (superseded by #41), #37 (rest landed in #40), #8, #15, #26, #28, #29. They can be closed with this one, as can #42–#51 and #1–#5.

Navigation

Sidebar section	Pages
Start here	Home
Step 1 · Set up and run	Upload data, Describe experiment, Run analysis
Step 2 · Main results	Run summary, Compare groups (new), Volcano plot, Identified proteins
Step 3 · Explore further	Sample similarity (PCA), GO enrichment, Clustered heatmap, Protein abundances, Per-sample counts, Heatmap
Export	Download results
Fine-tune the comparison	1. Filter proteins, 2. Missing values, 3. Normalize, 4. Statistical test
Quality control and details	Identification quality, Peptide-spectrum matches, Mass shifts, Rescored hits (Percolator), Raw search hits (Comet)

The sidebar shows the first ten pages (through GO enrichment) and folds the rest behind "View more", so the expert and QC pages are still there but less prominent.

Guidance on every page

• src/common/guide.py: page_intro (what the page answers, background in an expander), next_steps footer, needs_results for pages opened before a run.
• Home rewritten as a three-step guide with a "Which page answers my question?" table.
• Every page's text rewritten in plain words (PSM, FDR, fold change, adjusted p-value, PCA, GO terms, imputation explained where they appear).

Sensible defaults

• Compare groups (content/compare_groups.py, src/common/differential.py) runs filtering → imputation → normalization → statistics with the recommended settings, continuing from any step already applied on the Fine-tune pages. Volcano, PCA, heatmaps and GO enrichment compute it on demand instead of telling users to go through four pages first. The page states which direction a positive log2 fold change means (openms-insight subtracts the alphabetically first group).
• Describe experiment: samples and groups is the first tab, tabs are named by purpose, the 11 TMT tool tabs are regrouped into 6, and the help boxes are rewritten (two of them described the wrong tool).
• LFQ search defaults to Orbitrap HCD (high_res, 0.01 Da fragment tolerance) instead of ion trap settings; the demo data is Orbitrap HCD.
• All uploaded runs are preselected, and the demo files get their groups (UPS2_ratio01 / UPS2_ratio10), so a fresh workspace runs the demo end to end without input.

Fixes found on the way

• Volcano, PCA, heatmap and clustered heatmap stayed on "Loading..." forever (on main too): regenerate_cache=True rebuilt the cache on every rerun, so the component kept requesting data. Caches are now keyed on the plotted data (plot_cache in insight_tables.py).
• GO enrichment results were drawn only inside the button block and vanished on the next rerun; they are kept in session state and shown in paginated tables.
• Removed a debug st.write of the results path in the TMT configuration.

Testing

• pytest tests/ test_gui.py: 213 passed (6 new in tests/test_differential.py).
• pylint --errors-only: only the existing pyopenms E0611 notices.
• Ran the demo data end to end with OpenMS 3.5 TOPP tools (bioconda) through Upload → Configure (defaults only) → Run, then opened every page in headless Chromium. Volcano, PCA and both heatmaps render. The demo finds no significant proteins at the default cut-offs because the trimmed demo files contain no quantified UPS2 proteins.
• GO enrichment was not exercised (needs MyGene.info access).

Screenshots are in the project files under quantms-web-guided/.

---

Generated by Claude Code

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 57d63dee-c9b6-41ad-a7ce-dc99668c536b

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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.}