Repository navigation
Wet-lab practice track (bench simulator, lab measurement and biosafety packages) and Statistics 0.7.0 - #1
Merged
Merged
Conversation
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A deterministic replay engine (courselab.bench) and stateless API run practice simulations of bench procedures: the browser sends a seed, a mode and the action list; the server replays it and returns only what a person at the bench could see, with a checkpoint debrief when the run ends. Hidden faults, checkpoint rules and model parameters stay on the server. Protocols, each a JSON definition validated by a new schema and paired with a Python model: - micropipette gravimetric verification (water density per Tanaka et al. 2001; acceptance limits are labelled practice values) - decimal dilutions and the aerobic plate count, adapted from FDA BAM Chapter 3 (January 2026 edition) - a Class II biosafety cabinet session with a spill, adapted from BMBL 6th edition Appendix A Also adds a Bench practice page in the React app, registry records for the three sources, a content-validator check for bench protocols, and 51 backend and 2 frontend tests. Formative practice only; unreviewed. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…of the wet-lab track A new six-week foundation package, the catalog's 26th: concentration and solution preparation, dilution series and error propagation, volume measurement and gravimetric pipette checks, balances and minimum weight, pH-meter calibration and buffers, and spectrophotometry with standard curves. 64 practice items, 26 cards, a self-check case and an optional reagent and assay qualification project (guide, four rehearsal items, five Case studio milestones). Lessons 2 and 3 link the plate-count and pipette bench simulations. Scope compared by topic with MIT OCW 5.310 Laboratory Chemistry, Fall 2019 (link-only comparator record). The package sits in the tissue-regenerative-medicine pathway before the cell-culture node. A new package test recomputes every numeric key and caught two items whose trailing-zero keys (750, 100) the three-significant-figure grader would have rejected; both are now exact-value items. Count pins, the lesson-link test (now accepting #/bench/ routes that name a protocol), README, roadmap, audit, plan index and generated reports are updated. Partial, unreviewed and formative-only. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…rack A new six-week package, the catalog's 27th, adapting BMBL 6th edition (Sections II-IV, Appendices A, B, C, H) and OSHA 29 CFR 1910.1030, 1910.1450 and 1910.1200 (US federal works) with paragraph and page references: biological risk assessment, BSL-1/2 practices, biosafety cabinets, decontamination and sterilization with D-values and autoclave verification, human materials and bloodborne pathogens with exposure response, and chemical hygiene, hazard communication and transport. 64 practice items, 26 cards, a self-check case and an optional iPSC biosafety plan project with five Case studio milestones. Scope compared task by task with the ABSA CBSP examination content outline (2026, link-only); program-management and facility-review tasks are outside the package by design. New registry records for the three OSHA sections and the CBSP outline; count pins, docs and reports updated. Partial, unreviewed and formative-only; does not replace institutional safety training. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A fourth protocol (layer 1): prepare a typical waste load, place thermocouples and biological indicators, run a steam cycle, incubate the indicators with a positive control and report the centre heating lag, the verdict and the validated exposure time. Hidden heating lag and indicator D-values change from run to run; a sealed bag traps air so the centre never reaches temperature. Bagging, tray, indicator and thermocouple placement, 121 C and routine cycle records follow BMBL 6th edition Appendices A and B; other values are labelled practice or model values. Six new backend tests. Linked from lab-biosafety lesson 4 and its syllabus; README, roadmap, plan and a new docs/WET_LAB_LAYER_OUTLINES.md (layers 0-5) updated. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A fifth protocol (layer 1): a needlestick, cut or eye splash with primary human cells is chosen at random, and the learner must decontaminate at once by the right method, make the work safe, report immediately, contact occupational health, document the route and source material and complete the incident record. Checkpoints follow OSHA 29 CFR 1910.1030 (d)(2)(vi) and (f)(3) and BMBL BSL-2 special practice 7; the 15-minute eye flush and the 30-minute reporting window are labelled practice values. Six new backend tests. Linked from lab-biosafety lesson 5 and its syllabus; README, roadmap and layer outlines updated. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… and one-way ANOVA (0.7.0) A new scope crosswalk sets MIT OCW 18.05 Spring 2022 beside the University of South Carolina STAT 205 Fall 2011 schedule (link-only comparator record). Both teach chi-square methods and STAT 205 also teaches 2 x 2 tables, stratification and one-way ANOVA, which the package lacked. Two original readings: categorical outcomes in week 5 (risk difference, risk ratio and odds ratio with a log-scale interval, chi-square and Fisher's exact tests, Simpson's paradox and the Mantel-Haenszel odds ratio) and one-way analysis of variance in week 6 (sums of squares, F, eta squared, pooled-variance comparisons with a Bonferroni adjustment, ANOVA from summaries). 20 practice items and eight cards; all data synthetic. The crosswalk records that likelihood estimation and normal conjugacy, listed as limited in the older audit, were already taught. Logistic regression and ROC curves remain absent. Snapshot (0.6.0 archived), handoff, progress record, audit, plan index, roadmap, reports and version pins updated; the new items are added to the independent statistics recalculation table. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Seven commits, each verified before it was pushed. This is a personal hobby and learning project; all of this was developed with substantial AI assistance (Claude Code), and none of it has been reviewed by a laboratory professional, biosafety officer, statistician or certifying body. Every package stays
partial,unreviewedand formative-only.Wet-lab practice track
docs/WET_LAB_PRACTICE_PLAN.md,docs/WET_LAB_LAYER_OUTLINES.md): a layered track (measurement, biosafety, cell culture, molecular methods mapped to the ASCP MB outline, quality systems) and a bench-procedure simulator.backend/courselab/bench/,/api/v1/bench/..., a React Bench practice page): stateless, deterministic replay from a seed and the learner's action list. Hidden faults and checkpoint rules stay on the server; a checkpoint debrief is returned when a run ends. Five protocols:Statistics 0.7.0
Second-comparator crosswalk (MIT 18.05 Spring 2022 and USC STAT 205 Fall 2011) with two new readings: categorical outcomes (2 × 2 tables, risk and odds ratios, χ² and Fisher's exact tests, Simpson's paradox and Mantel–Haenszel) and one-way ANOVA. Logistic regression and ROC curves remain listed gaps.
Sources and rights
US federal works (BMBL, FDA BAM, OSHA) are adapted with attribution; ASCP, ABSA, ISSCR, MIT OCW and USC material is link-only and sets scope only. Registry records were added for each, and cited identifiers were checked with the citation-check script.
Verification
Run in the Linux dev container with
.claude/skills/courselab-dev/scripts/verify.sh --frontendfor each batch. Last full run: content validator 0 errors (27 packages); root pytest 4,673 passed with one failure, fixed afterwards (the new statistics items' recalculation entries and tolerances; the affected 1,054 tests then passed); backend pytest 8,266 passed; security scan 0 findings; ruff clean; frontend lint, 84 unit tests, build and bundle scan passed (one flaky frontend test failed once and passed on rerun).Not run: the Docker Compose stack and the Playwright end-to-end suite, which need the host or CI.
🤖 Generated with Claude Code