Skip to content

Add schema support for phosphoproteomics and new dataset outputs - #491

Merged
sgosline merged 3 commits into
build-orchestration-and-safetyfrom
schema-phosphoproteomics
Sep 29, 2026
Merged

sgosline merged 3 commits into
build-orchestration-and-safetyfrom
schema-phosphoproteomics

Conversation

@jjacobson95

@jjacobson95 jjacobson95 commented Sep 21, 2026 •

Copy link
Copy Markdown
Collaborator

Pipeline Hardening/Debugging PR # 13

Add schema support for phosphoproteomics and new dataset outputs

Extends the CoderData schema so the new phosphoproteomics data type, the phosphosite reference, and the single-dose viability metric are all valid, and declares the expected output files for the new datasets. Must merge after the dataset PRs that produce these outputs, or schema validation will fail on files that don't exist yet.

Data model additions (schema/coderdata.yaml)

  • phosphosite_id: a new integer identifier slot for a phosphorylation site (gene + residue + position).
  • Phosphosite class: the reference table produced by the phosphosites builder; each row is a unique gene/residue/position with entrez_id, gene_symbol, residue, position, and modification.
  • Phosphoproteomics class: per-sample phosphoproteomics measurements, keyed by phosphosite_id and improve_sample_id, storing the log-normalized, batch-corrected abundance value.
  • uM_viability added to the ResponseMetric enum (single-dose viability fraction at 1 μM), so the cNF single-concentration experiment values validate.

Expected-file declarations (schema/expected_files.yaml)

  • Declares the cnf outputs (samples, transcriptomics, proteomics, phosphoproteomics, experiments, drugs, drug descriptors), with mutations and copy number left commented out since cNF has none.
  • Declares the phosphosites reference output and the genes reference output.
  • Adds an mpnst Combinations target and the novartis drug-descriptor file to keep the expected-file list in sync with the current builders.

Merge-order requirement

  • Must merge after the phosphosites, cnf, and MPNST PRs, because expected_files.yaml references outputs those PRs generate; landing this first would make the schema check fail.

Scope: 2 files. Base: build-orchestration-and-safety.

Related issues

Closes #469

@jjacobson95
jjacobson95 force-pushed the build-orchestration-and-safety branch from cfa266d to 5be9072 Compare September 21, 2026 21:55
@jjacobson95
jjacobson95 force-pushed the schema-phosphoproteomics branch from 1d6624f to 31d396c Compare September 21, 2026 21:55
@jjacobson95 jjacobson95 added the Schema update Change or addition to the schema label Sep 22, 2026
@jjacobson95 jjacobson95 modified the milestones: 2.3 new build, 2.4 new build Sep 22, 2026
@sgosline

Copy link
Copy Markdown
Member

We are not including cNF and phosphosite in this build

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I am guessing that the same as with the other issues that contain CNF applies here as well, i.e. we keep the code but don't include the CNF data in the bulid?

@sgosline
sgosline merged commit b5518e1 into build-orchestration-and-safety Sep 29, 2026
@sgosline
sgosline deleted the schema-phosphoproteomics branch November 13, 2026 00:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Schema update Change or addition to the schema

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants