{"resourceId":"utoronto-openpep-google-org-open-peptide-binding-data-20261016","versions":[{"version":"external-511f55fefb3c62970fe3ff7d2ae9225ef29144a77b5b2d0e16b84db800ee5dd8","resource":{"id":"utoronto-openpep-google-org-open-peptide-binding-data-20261016","title":"University of Toronto launches OpenPep with US$2.3M from Google.org to generate open peptide-binding data and train an open AI model","organization":"University of Toronto (Leslie Dan Faculty of Pharmacy, Donnelly Centre, Acceleration Consortium); Google.org","sector":"Research university (pharmaceutical sciences)","geography":"Canada (transferable to U.S. research universities pursuing funder-backed open datasets)","publishedAt":"16 October 2026","publicationDate":"2026-10-16","eventDate":null,"sourceName":"University of Toronto News","sourceLabel":"U of T News, \"AI-powered OpenPep platform to accelerate peptide drug discovery\", 16 October 2026","sourceUrl":"https://www.utoronto.ca/news/ai-powered-openpep-platform-accelerate-peptide-drug-discovery","evidenceClass":"independent-reporting","outcomeClass":"emerging","topics":["knowledge-work","governance-procurement","operating-model"],"finding":"U of T News reported on 16 October 2026 that a team led by Joseph Brown (Leslie Dan Faculty of Pharmacy) received US$2.3 million through the Google.org Impact Challenge: AI for Science for OpenPep, an open-access peptide database and AI model. Keith Pardee's group will produce up to 500 disease-relevant proteins with cell-free systems; Brown's team will test peptide libraries against them by mass spectrometry; the data will train models predicting peptide-protein binding, with validated predictions fed back. The team will crowd-source target proteins from hospital and other collaborators. The rationale is a data gap: peptide discovery lacks the large datasets other drug-discovery AI relies on. No timeline, licence, release schedule or performance target is given.","sledRelevance":"Shows a funder paying a university to generate the training data first, then build an open model, rather than funding a model on existing data. Research offices weighing AI-for-science grants can use it as a template for data-generation-led projects with an open-access commitment.","evidence":"Opened the University of Toronto news article dated 16 October 2026 and read the full text, including funding, project design, team roles and crowd-sourcing plans.","architectureImplications":"Requires a pipeline from lab instruments (mass spectrometry) to a curated, versioned database and model training, with a loop for adding validated predictions.","governanceImplications":"Open-access commitment needs a defined licence, data standards and release cadence; crowd-sourced targets from hospitals need intake rules.","securityPrivacyImplications":"Protein and peptide data are not personal data; collaborator target suggestions may carry confidentiality or IP expectations.","caveats":"University news item about a newly funded project; no results, timeline, licence or model benchmarks yet. Funding is corporate philanthropy, and terms with Google.org are not described.","streamIds":["research"],"roles":{"sales":"Lead with data generation as the asset. OpenPep is funded to produce new peptide-binding data first and build an open model on it. Research VPs, sponsored programs and lab-automation leaders can use this framing for AI-for-science proposals. A bounded engagement helps design the data pipeline, licence and release plan for a funded project. Interpretation: the project is just funded, so it supports no claims about model accuracy or faster drug discovery.","engineering":"Treat this as a lab-to-model data pipeline. Mass spectrometry output must become a curated, versioned database feeding model training, with validated predictions looping back. Pre-sales should check instrument data capture, metadata standards, storage, training compute, and how an open release will be hosted, versioned and cited by other labs. Interpretation; the article gives no technical design, licence or hosting plan.","delivery":"Frame delivery as standing up an open research data service. Work would cover data standards, curation, versioned releases, a licence, and an intake process for crowd-sourced targets. The lead lab owns it with research data services and sponsored programs, and someone must be named to host the database after the grant. Interpretation: acceptance should rest on data released and reused, because the source reports plans, not outcomes."},"roleDetails":{"sales":{"customerProblem":"AI-for-science projects in data-poor fields need a credible plan to generate, curate and openly release training data.","stakeholders":["Principal investigator","Vice president for research","Sponsored programs office","Research data services or library"],"discoveryQuestions":["What licence and release cadence does the funder expect?","Who will host the open database after the grant?","How are instrument data captured and annotated today?"],"valueHypothesis":"A clear data-pipeline and release plan may strengthen funded AI-for-science projects and their reuse.","potentialEngagement":"A four-week data pipeline and open-release design for one funded project, covering standards, licence and hosting.","unsupportedClaims":["Model accuracy or binding-prediction performance","Faster drug discovery","Specific release dates"]},"engineering":{"fit":"Fits projects that generate structured lab data at scale for model training; limited where data are sparse or proprietary.","architectureIntegration":"Instrument data capture into a curated, versioned database feeding training jobs, with feedback of validated predictions.","prerequisites":["Metadata and data standards for assays","Storage and versioning platform","Model training compute"],"constraints":["Licence and terms not yet public","Long-term hosting unfunded beyond the grant","Crowd-sourced targets may carry IP limits"],"security":"Protect collaborator-submitted targets and any pre-publication data until release; apply access control to staging areas.","proofOfValue":["Ingest one assay batch end to end into a versioned release","Validate that a released dataset reproduces a published model split","Test intake of external target suggestions with IP screening"]},"delivery":{"work":"Define data standards, curation workflow, release versions, licence and target intake process.","dependencies":["Funder agreement on licence","Research data services capacity","Lab data capture tooling"],"ownership":"Lead lab with research data services and sponsored programs.","skillsAdoption":"Data curation and metadata training for lab staff; ML team access to versioned data.","governanceCheckpoints":["Licence approval before first release","Release review for each version","Annual hosting and sustainability review"],"acceptanceCriteria":["Proposal: first public data release with a defined licence within the grant period","Proposal: every release versioned with documented assay metadata","Proposal: external downloads or reuse tracked each quarter"],"risks":["Open-release commitment slipping","No host after grant end","Data quality variation across batches"]}},"retrievedAt":"2026-10-19T21:58:32-05:00","enrichedAt":"2026-10-19T21:58:32-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Open data could lower entry barriers for smaller labs; the project needs data engineering and curation skills alongside wet-lab work.","procurementImplications":"Not a procurement; grant terms (licence, publication, any platform expectations) should be checked by sponsored programs.","operatingModelImplications":"Pairs a wet-lab data factory with an ML team and a self-driving-lab initiative (Acceleration Consortium); long-term hosting of the open database needs an owner.","sourceVerification":{"openedUrl":"https://www.utoronto.ca/news/ai-powered-openpep-platform-accelerate-peptide-drug-discovery","referenceExcerpt":"Pardee and his colleagues will draw on their expertise in accessible cell-free systems from synthetic biology to produce up to 500 disease-relevant proteins","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}