Learn Fifty-five original destinations implement the finite SC14 Microbiology scope: authentic source-labeled microscopy, growth/metabolism, genetics, ecology/applications, host evidence, control interpretation and whole-course synthesis. Synthetic models and public solutions remain distinct from actual observations and independent performance.
This is the course outline. Open a unit to choose a lesson; every lesson opens on its own page. Explanations and experiments live inside those lesson pages.
Preparation: General cell biology, membranes, basic chemistry, ratios and graph interpretation. No biological handling, clinical training or culture experience required.
Course outcome: Compare microbial structures and microscopy evidence using calibration, geometry, source identity and explicit limits on morphology/counting; later modules extend to population, genetic, ecological and host/control reasoning.
All lessons are directly accessible. Reused foundation explanations are identified where relevant; completing a different course does not award this course's evidence. The 55 registered lessons keep separate predictions, captures, practice responses and reflections.
Compare representative microbial structures, calibrate and interpret image measurements, audit actual public-domain micrographs and bound identification/counting by morphology, resolution and detection.
Compare unresolved, nucleus-positive and peptidoglycan-positive packets, then add ribosome or chloroplast evidence. Retain each candidate set and explain an empty intersection using an observation or candidate-set limitation.
Reconstruct both endpoint ratios before changing calibration span or object size. Compare relative uncertainty, distinguish bounded error from confidence and explain why the actual unlabeled micrographs cannot inherit this fictional scale.
Compare wavelength and numerical aperture separately. Calculate d, retain the equality convention and explain an acquisition or contrast limitation outside the ideal criterion. Relate the DIC actual image to contrast rather than literal surface height.
Change offset within section mode and then switch to projection. Retain the three-dimensional assumption and explain why missing a structure in one plane is not evidence that the whole cell lacks it.
Compare L=0,2,4 at a fixed radius, then change radius. Reconstruct cap and cylinder contributions separately and explain why morphology and surface-to-volume ratio cannot by themselves identify a species or predict growth.
Vary each coefficient separately and preserve the intermediate concentration and both drops. Compare the serial formula with a mistaken parallel sum, then identify a real envelope feature or active process that the passive model omits.
Inspect both actual images and retain two visible features, one unresolved claim and exact source identity. Then compare both counting rules on the separate original schematic, including and excluding the ambiguous feature. Explain why neither actual species nor viable count follows from that schematic.
Hold true composition fixed while changing detection, then hold detection fixed while changing composition. Retain both denominators, distinguish expected counts from actual observations and propose a validation or sampling observation needed to interpret a real field.
Analyze supplied population and recoverable-unit records, conserve metabolic and resource budgets, and separate environmental mechanisms from detection and mixture effects without culture protocols.
Reconstruct each boundary and selected observation, then vary the fictional doubling interval. Compare a within-phase logarithmic slope with a whole-record endpoint slope, and propose an independent observation that could test the assumed phase change.
Compare event combinations giving the same net count and preserve gross daughters, replaced parents and losses separately. Identify what a second endpoint still cannot reveal and propose a nonoperational evidence type that would separate turnover components.
Vary separation and recovery independently. Retain the fixed 14 cells, expected unit denominator and operational measurement boundary. Explain what extra information would be needed before converting an actual recoverable-unit count to total viable cells.
Reconstruct the equivalent volume before calculating concentration and bounds. Change the archived exponent and count separately; explain the tenfold scaling and which uncertainty sources remain omitted. Use supplied numbers only.
Hold N fixed while changing background, then hold background fixed while changing N. Retain the inverse domain, compare nominal and biased estimates and explain why optical abundance is not a recovered viable-unit count.
Compare all-respiration, all-fermentation and mixed formal records. Close carbon and electron ledgers separately and explain which additional observations would be needed to infer actual biomass or energy yield.
Compare a maintenance-limited and a separate-nutrient-limited record. Close maintenance, growth use and unused inventory, retain unmet demand explicitly and explain why the toy priority rule does not establish a real cultivation strategy.
Compare crude and fixed-standard contrasts for several mixture pairs. Retain both within-category effects and the interaction; name a design or evidence limitation that prevents treating the standardized number as an automatic causal effect.
Change diffusion and consumption separately. Reconstruct both boundary fluxes, their difference and the midpoint concentration. Explain why a measured gradient without independent source/uptake information cannot uniquely identify metabolism or an organism.
Trace copying and expression, distinguish horizontal-route evidence from lineage association, and interpret preexisting variation, sorting and sequence uncertainty without biological manipulation.
Reconstruct duplex and strand totals, then change allocation without changing copying rounds. Separate the conservation statement from a claim about equal daughter inheritance and identify what the ledger does not locate.
Find two supported settings with the same abundance but different flux. Reconstruct production and removal separately, then explain why a snapshot cannot establish their individual values.
Compare versions at a fixed offset before changing the offset itself. Retain exact triplet boundaries and first stop, then distinguish a coding consequence from an unsupported phenotype claim.
Reconstruct absence and detection probabilities across supported sample sizes. Name the sampling unit and one correlation or selection bias that would invalidate the product rule.
Compare an unknown clue, a repeated compatible clue and a conflicting clue. State the exclusive-event assumption and propose a nonprocedural alternative interpretation such as multiple events or an unreliable initial classification.
Construct two records with the same pooled total and different lineage allocation. Retain each denominator and explain why neither pattern alone uniquely proves transfer or clonal expansion.
Reconstruct the descendant tree and descriptive moments for early and late events. State which timing and inheritance rules were stipulated, and name an alternative source of variation in an actual family record.
Change the two stages separately and reconstruct both denominators. Identify whether a difference arises before or after normalization; do not attribute the final result to a new mutation that the model excludes.
Recompute both event likelihoods and change the prior without changing read evidence. Explain one dependence or omitted-model limitation that could make this numerical posterior unsuitable for an actual record.
Interpret community composition with observation bias, conserve cross-feeding and nutrient currencies, and evaluate functional/application claims within explicit transport, material and causal limits.
Reconstruct counts and fractions for a fixed-A record and an equal-scaling record. Identify which conclusions require an independently measured total and which follow from composition alone.
Compare equal and unequal recovery while keeping the underlying community fixed. Reconstruct each denominator, then state what would be unknowable if the factors were not independently supplied.
Draw donor, receiver and combined boundaries. Reconstruct all carbon terms for two allocations, then identify which energy or mechanism claims cannot be obtained from an atom ledger alone.
Reconstruct all nitrogen pools before and after the second transformation. Separate aqueous retention from gas export, then name one omitted intermediate or coupled resource that prevents a complete ecosystem prediction.
Compare capacity, uptake and base concentration as one mathematical parameter changes. Check the zero-gradient base boundary and identify what physical or biological evidence would be missing for a real biofilm interpretation.
Write the denominator next to every yield. Reconstruct all endpoint carbon pools and name a downstream recovery or quality boundary that the record does not cover.
Compare the two equal-function records and then remove A mathematically while holding B fixed. Identify which conclusions depend on additivity and which would require independent compensation or interaction evidence.
Reconstruct both site changes and both contrasts. Describe a plausible pretrend or concurrent-event alternative and an additional documented observation that could distinguish it without prescribing field or biological handling.
Compare physical/innate/adaptive mechanisms and interpret fictional host and population records without equating detection, response, contact or association with disease causation or protection.
Change the classification artifact while holding both true pools fixed. Reconstruct total conservation and both denominators, then name an independent evidence requirement before making an invasion or disease claim.
Close arrival, removal and accumulation at one time and at steady state. Compare records with the same steady stock but different time responses, separating mechanism assumptions from actual host evidence.
Compare independent overlap with its allowed dependence bounds. Describe a shared-cause alternative and distinguish an event-recognition record from a measured protective outcome.
Reconstruct each old-stock removal and new arrival. Compare equal endpoint stock with unequal throughput, then explain why the cumulative stock alone cannot establish protection or injury.
Hold class fixed while changing each supplied condition, then change class without changing conditions. Separate canonical recognition context, evidence of matching and the omitted biological response requirements.
Compare timing, peak and contraction at the same observation time. Reconstruct a value from the appropriate interval and name an independent functional endpoint needed before interpreting protection.
Construct the full two-by-two table before dividing. Change only prevalence, then only specificity, and preserve the denominator and population assumptions for every comparison.
Hold new events fixed while changing resolution. Reconstruct initial-risk, total-population and ever-affected denominators, then identify why person-time incidence cannot be recovered from these totals alone.
Reconstruct both crude mixtures and the fixed-standard comparison. Name the target population and one unmeasured factor that would keep this from being a randomized causal estimate.
Replay the contacts in chronological order and retain the earliest possible arrival at each node. Compare the disputed-edge scenarios, then explain what source or host evidence would be needed beyond reachability.
Compare control mechanisms and fictional susceptibility evidence with explicit recovery, censoring, phenotype, source-frame and validation limits; design a virtual discriminating comparison without biological execution.
Draw surface and overall boundaries, reconstruct all pools and distinguish material removal from a measurement of recoverability. Identify which additional claim would require a validated process rather than arithmetic.
Reconstruct both component contributions before taking a logarithm. Compare successive log changes and name a different measurement explanation that could also produce a curved record.
Write the observation inequality before transforming it. Compare two reporting limits at the same initial count, preserving the direction and strictness of the log bound.
Hold r fixed and change q, then distinguish physical change from measurement change. Explain what independent calibration evidence is required before dividing by a recovery factor.
Reconstruct the geometry and compare two source sizes at the same outer radius. Identify which additional biological and calibration evidence would be required before any susceptibility interpretation.
Reconstruct both component contributions and compare the same count under two supplied phenotype-history records. Name which growth-threshold or survival distinction must be established independently rather than inferred from the count.
Reconstruct both denominators and compare two equal-total records. State which evidence would distinguish mixture change, reversible expression and an unmeasured new genetic origin.
Compare unique and entry-weighted fractions while changing duplicates only. Then change sample mixture and explain why a standardized comparison is not the same as evidence of representativeness.
Calculate all candidate residuals, preserve ties and state the unused predictions before revealing a new record. Explain how uncertainty or a value outside every prediction would change the conclusion.
Integrate all six capability modules while preserving source identity, measurement currencies, competing mechanisms, first reasoning and independent/practical validation boundaries.
Reconstruct the true and observed records separately, then vary recovery without changing biology. Keep expression and carbon currencies distinct and propose a source-aware additional observation needed before a causal host or control claim.
Reconstruct both contingency expectations, separate mixture from observation changes and retain a competing causal explanation. Complete the retained full-course evidence stages without treating public solutions, actual source labels or an automated status as independent mastery.
Keep predictions before comparison, complete raw records and units, explain competing models and test a changed case. Supplied and simulated data are labeled in each lesson. A saved notebook is not proof of experimental technique or correct reasoning.
Evidence to retain: Retain positive-versus-unresolved features, all compatible candidates, calibration endpoint intervals, optical criterion and section/projection predictions before feedback.
Boundary: Original measurements and diagrams; no acquisition or unknown biological material handling. Actual photographs cannot inherit the fictional calibration.
Evidence to retain: Close area/volume and serial-flux quantities; audit both actual images with visible features and source identity; preserve count rule, ambiguous features and detection-weighted versus underlying proportions.
Boundary: Source-labeled public-domain images, separate original count diagram and fictional detection expectations. No species, viable-count, culture or clinical susceptibility inference.
Evidence to retain: Retain phase-specific slopes, gross/net event ledgers, aggregate/recovery expectations, original-equivalent volumes and nonlinear blank corrections. Explain which distinct quantity each method measures.
Boundary: Original archived fictional records only; no dilution preparation, plating, incubation, culture or viability/clinical claim.
Evidence to retain: Close carbon/electron endpoints and maintenance/growth/unused substrate separately; retain unmet demand and an independent resource cap.
Boundary: Formal ensemble amounts and fictional coefficients, not actual biomass/ATP yield or process optimization.
Evidence to retain: Compare within-stratum and standardized effects, then close boundary flux difference against distributed consumption. Preserve alternative mechanisms and missing sampling evidence.
Boundary: Coded factors and original mathematical layer, not organism-specific environmental conditions, gradient-generation methods or causal species identification.
Evidence to retain: Retain total/original strand ledgers, unequal pools, abundance versus flux and exact partial-code frame interpretations. Distinguish coding effects from biological phenotype.
Boundary: Original mathematical records and nonfunctional token strings; no genetic engineering or measured expression protocol.
Evidence to retain: Keep sampling units, no-variant probabilities, route candidate intersections and both within-lineage denominators; identify competing histories.
Boundary: Interpreted fictional evidence, not a method to obtain or transfer biological material; no unique mechanism from association.
Evidence to retain: Reconstruct inherited clusters, finite-family variance, sequential sorting/loss and likelihood/prior/posterior distinction with independence limitations.
Boundary: No mutation-rate estimate, treatment or resistance induction; independent practical competence remains separate.
Evidence to retain: Reconstruct absolute and relative counts, recovery-weighted observations and the independently assumed inverse; preserve sampling/measurement unit and missing calibration.
Boundary: Original fictional class counts; no actual extraction method or species composition claim.
Evidence to retain: Close carbon and nitrogen atom budgets, distinguish internal transfers, integrate a reactive profile and compare product yield denominators.
Boundary: No living system operation, process optimization, actual biomass yield, energy closure or quality guarantee.
Evidence to retain: Compare equal-function communities and removal dependence, then retain separate carbon/nitrogen change contrasts and competing explanations.
Boundary: Original descriptive records; no actual field intervention, automatic compensation or causal/effluent-safety certification.
Evidence to retain: Keep true/apparent compartments, conserved arrival/removal, channel-dependence bounds and simultaneous recruitment/resolution distinct from causal disease or damage claims.
Boundary: Fictional nonclinical units; no participant sampling, pathogen exposure, diagnosis or treatment.
Evidence to retain: Trace canonical peptide/MHC context and limited response gates, reconstruct staged proxy kinetics and full population test denominators.
Boundary: No actual vaccine efficacy, individual immune status, complete activation or patient test interpretation inferred.
Evidence to retain: Separate incidence/prevalence, preserve stratum weights and replay chronological contacts before assigning possible paths; retain missing counterfactual/source evidence.
Boundary: Anonymous fictional cohorts/networks, not real contact tracing, outbreak reconstruction or intervention efficacy.
Evidence to retain: Retain material versus recoverability pools, component contributions/log increments and strict endpoint bounds without inventing zero or a validated sterile state.
Boundary: Fictional abstract records; no temperature, dose, duration, culture or control-process execution.
Evidence to retain: Compare independent recovery corrections, source-disk/annulus geometry, endpoint versus susceptibility/history evidence and aggregate versus per-carrier expression.
Boundary: No live assay, clinical susceptibility classification, manipulation or generation/selection of resistant organisms.
Evidence to retain: Audit unique versus duplicate entries and standardization; preserve candidate ties and unused predictions in a virtual discriminating plan.
Boundary: No actual prescribing rule, claimed surveillance prevalence or physical experiment authorization.
Evidence to retain: Reconstruct true versus recovered composition, marker carrier versus expression denominators and independent carbon budgets; perform a separate actual licensed image audit.
Boundary: Original numerical records are not the source-image population or actual host/clinical evidence; source labels and public feedback remain exposed guidance.
Evidence to retain: Separate population mixture and classification change, then retain a source-aware whole-course plan/report/critique/revision/unfamiliar response with prior category archives.
Boundary: Authored material is not completed learner mastery, actual wet-laboratory performance, clinical authority or institutional equivalence.
Open the retained response and project workspaces. Practice, saved work and independently reviewed proficiency remain separate.
These 38 course-specific public tasks are practice, not a secure examination. Each retains up to three immutable written responses, known feedback exposure before each response, and human rubric critiques. Drafts remain editable. No text-length check or self-reported review certifies correctness, independence, laboratory skill or mastery.
Open both actual linked public-domain micrographs and provenance. Record creator/source/version/hash, reuse, visible features and an alternative interpretation; distinguish source species labels from your own observation. Neither photo has an embedded calibrated scale. Separately a fictional 10µm reference spans 200±2pixels and an object 40±1pixels. Derive its nominal length and hard endpoints without attaching them to the actual photos.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Original population stays 100 through time 2, doubles every 1 through time 6, then halves every 2 through time 8. Find peak/final. Separately 3mol glucose is split one half complete respiration and one half formal ethanol fermentation. Find CO2, ethanol, O2 and carbon/electron closure; omit any universal ATP claim.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
One duplex completes four copying rounds then pools shift two from an equal split. Count original/new strands and pools. Separately four independent reads have at least three variant calls; error-only probability 0.2 per read, alternative 0.5 and equal model priors. Calculate the posterior and state why it cannot establish a phenotype or new need-induced origin.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Carbon input 24, donor biomass fraction 0.5, half remainder to CO2/half intermediate; receiver uses 40% intermediate C as biomass,40% as CO2, leaves 20%. Close endpoints. Compare community A 10/B 20/C 40 with contributions 3/4/0 against A 30/B 20/C 40 with 3/1/0. What differs despite equal initial function?
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
True low/high state frequencies 0.1/0.4 remain unchanged. First record half/half, sensitivity 0.8, specificity 0.9. Second high fraction 0.8, specificity 0.8, same sensitivity. Compute true mixtures, classified positives, apparent change and second under first measurement properties. Does the change establish disease worsening or failed control?
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Initial 10000 with endpoint strictly below 2: state the positive-endpoint log bound. Separately initial 1000 with slow fraction 0.2, fast/slow retained fractions 0.01/0.5: find remaining and slow share, and name additional phenotype evidence needed. Candidate B[2,2,1] and C[2,2,3] fit observed[2,2]; define a discriminating unused record without an experimental execution protocol.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
True A 400/B 600, marker only A, A recovery 0.75, B recovery 0.5, independent expression 3 per true A and carbon uptake 2 per A/1 per B. Reconstruct observed composition, expression denominators and naive/corrected carbon. Connect an actual image audit and all six modules to a first report, actual critique, justified revision and unfamiliar response without claiming completed practical or clinical competence.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Original100 units have removal0.6 and retained recoverability0.2. Find all pools. Separately S(x)=0.9(0.1)^x+0.1(0.8)^x at index3: find S and slow share. A third packet has initial1000 and endpoint strictly below3; state the positive-endpoint log-reduction bound.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Initial1000,true remaining fraction0.1 and independently supplied recovery0.25: find measured count and naive/corrected log reduction. Separately original clearing-image outer radius4/source radius1.5: find diameter,width and annulus area. Explain why these do not establish clinical potency or killing.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Original1000 units have slow fraction0.1, fast retention0.01 and slow retention0.5. Find contributions and slow share. Separately compare marker fraction0.2/expression3 with0.6/expression1. What additional evidence distinguishes resistance/tolerance or carrier expansion/expression change?
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Unique100 records have high fraction0.2, low/high state fractions0.1/0.4 and each high record appears three times. Find unique and entry-weighted fractions. Separately candidates A[1,1,1],B[2,2,1],C[2,2,3],D[3,1,2] fit observed first pair[2,2]. Identify ties and unused predictions.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Select two competing interpretations of a supplied control record. Define an unused documented observation, specificity/recovery assumption, independent unit, source/permission and uncertainty criterion that would distinguish them. Preserve first predictions. No culture, biological acquisition, chemical preparation, antimicrobial exposure, resistance generation or clinical decision.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
True surface60/deeper10 units have20% of surface misassigned deeper. Find apparent pools and artifact share. Separately dX/dt=10−X,X0=0: find steady stock, time-one stock, clearance and accumulation. Explain why neither record establishes infection.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Recognition probabilities0.2/0.6 are stipulated independent. Find both/either/neither and overlap bounds if dependence is unknown. Separately R0=8,Rnext=0.5R+2 for four steps: reconstruct stocks and total resolved. State why neither number is a protection or damage measure.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Compare canonical MHC I with costimulation only versus receptor match plus costimulation in the simplified naive-cell gate. Then compare time-eight proxy2/4 for no-memory/memory records whose peaks are4/8. What can and cannot be inferred?
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Fictional1000-record test population has prevalence0.3,sensitivity0.8,specificity0.9. Reconstruct its table and positive predictive fraction. Separately a200-record cohort begins20 affected, gains20 new states among180 unaffected and resolves20 initial states. Find final prevalence and incidence proportion.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Use A low/high risks0.1/0.4 with high fraction0.2, B risks0.05/0.3 with high fraction0.8. Compare crude and half/half standardized differences. Separately initial source A at0 has A→B1,C→D2,B→C3; compare temporal reachability with and without A→C1.5.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Choose detection versus causation, immune proxy versus protection, or contact versus transmission. Using existing or fictional records only, state competing explanations, a discriminating documented observation, independent unit, denominator and source/privacy limitations. No participant testing, sample acquisition, exposure or treatment proposal.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Baseline A20/B30/C50 changes by A×1 and B,C×0.5. Find total and A fraction/absolute change. Separately, true A30/B70 undergo independently supplied recovery0.75/0.25; find observed and corrected A fractions. State why these are different packets.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Input12 mol C; donor biomass fraction0.25, half remainder to carbon dioxide and half intermediate. Receiver retains40%, respires40%, leaves20% of intermediate carbon. Separately100 mol N ammonium has80 converted to nitrate, of which75% becomes N2. Close both records without mixing molecules and atoms.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
For original layer L0.5,k1,D1,surface concentration1 and impermeable base, find capacity, uptake tanh(0.5), effectiveness and base1/cosh(0.5). Separately supply100 mol C, consume80%, route75% of consumed C to product and20% to biomass. Close final pools and both yield denominators.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
A contributes3 per abundance, B abundance20 with activity r, C abundance40 contributes zero to this function. Compare A10,r4 with A30,r1. Separately treatment carbon30→50/N export40→20 and reference carbon20→30/N export30→25: compute both change contrasts and state causal limits.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Choose a community-state versus observation-bias question or two explanations for a functional contrast. Propose a comparison using supplied or properly sourced existing records, define independent units/currencies, preserve missing values and predict a discriminating pattern. No field collection, biological handling or process setup.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
One duplex undergoes four complete copying rounds, then pools differ from an equal split by two duplexes. Find total/new/original strands and both pools. Compare steady dM/dt=20−2M with dM/dt=10−M; both use dP/dt=3M−2P. What is equal and what differs?
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Use partial code AUG=Met, GCU/GCC=Ala, GUU=Val, UAA=Stop. Compare AUGGCUUAA with AUGGCCUAA at offset zero and AUGUAAUAA. Separately, a population has supplied variant frequency0.1 and six independent cell draws; find no-variant probability and state what a negative sample cannot prove.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Two lineages have100 cells each. Compare carriers40/0 with20/20. Then one of eight16-cell endpoint families has a neutral event after generation two of four binary divisions. Find affected-family descendants, mean and population variance. Explain what these descriptive records do not establish.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Initial symbolic marker fraction0.4, carrier relative contribution1.5 versus noncarrier1, then marker retention0.75. Find pre/post-loss fractions. Separately observe at least three variant calls among four independent reads; error-only call probability0.2, alternative0.5 and equal prior probabilities. Compute the two-model posterior and its limitations.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Choose two competing explanations for one supplied record: hidden standing variation versus new origin, transfer versus lineage expansion, or production change versus altered removal. Specify a genuinely discriminating additional documented observation, sampling unit, predicted contrast and one remaining confounder. Do not propose culturing, genetic modification, transfer execution or resistance generation.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
In the original record, population stays100 through time2, then doubles every1 time unit through time6, then has half-life2 through time8. Find peak/final values. Separately compare100 starting cells with division/loss events60/50 versus40/30. Explain which histories a final count cannot identify.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
For the supplied14 viable cells in8 aggregates, use separation probability0.5 and recovery0.75. Separately use archived count40, fraction10^−2, analyzed0.2 mL and count-reading bound±5. Finally N50,B0 in the optical law B+2N/(100+N) is incorrectly blank-subtracted by0.1. Calculate all three results without calling them interchangeable cell counts.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Use3 mol glucose entirely through glucose→2ethanol+2CO2. Find carbon and electron endpoints. Independently use substrate30, maintenance rate4 for duration2, biomass yield0.5 and another nutrient cap6. Calculate maintenance,growth substrate,biomass and unused substrate,then explain why these formal records are not universal real yields.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
For condition responses A low/high2/4,B5/3,use high fractions0.8 at both conditions. Compare crude and half/half standardized contrasts and within-category effects. Separately use D1,q3 in the original unit layer C=10(1−x)−q x(1−x)/(2D). Find both boundary fluxes and midpoint. Name an alternative explanation that prevents a causal species inference.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Before feedback, choose an unfamiliar population or environmental record. Define what was counted or sensed,the original-sample denominator,source identity and units. Propose two mechanisms consistent with an endpoint,one discriminating observation,and a conservation or interval check. Specify uncertainty,replication and the no-biological-handling boundary.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Use exact 10 μm reference spanning 100±2 original pixels and object span 20±1. Find nominal length and interval. Independently use wavelength 600 nm, NA 1.5 and source separation 300 nm under d=0.61λ/NA. Explain why a larger display and either actual uncalibrated photograph cannot substitute for these supplied records.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
A sphere radius 2 μm is sectioned 1.5 μm from its center. Find section diameter and compare full projection. Separately a spherocylinder has radius 0.5 μm and straight length 0: calculate surface, volume and ratio. Explain why neither geometric result proves taxonomy or exchange rate.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
For serial permeability coefficients 3 and 1 with outside 12/inside 0, find shared flux and intermediate concentration. Separately a 1000-unit pool has trueA fraction 0.8, Adetection 0.25 andBdetection 0.5. Find expected detected counts and their composition; distinguish a ratio of expectations from an exact expected finite-sample ratio.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Inspect both bundled public-domain micrographs before using their descriptive labels as evidence. Retain file identity/hash, creator, source/reuse declaration, two visible features and at least one unresolved claim for each. Distinguish the actual photograph from the separate original object-count diagram. Explain why no calibrated physical size or viability follows and declare guided-label exposure.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Before feedback, plan interpretation of an unfamiliar supplied image packet. Define the object and counting rule, positive versus unresolved structures, calibration/contrast information, alternative explanations and a discriminating observation. Identify preparation/field versus biological replication and record actual-versus-original source identity. No acquisition, culture or staining protocol is requested.
Include calculations with units, source or dataset provenance, assumptions, competing explanations and limitations as relevant. The draft is saved but is not a locked response until submitted.
0/3 responses retained. No recorded feedback exposure for this task version. Earlier versions may already be familiar. Public inspectability still prevents certification of unfamiliarity.
Reviewer identity and role are declarations, not authenticated credentials. Unsaved review fields below are lost on navigation. Review every criterion with an exact quotation, calculation or evidence citation and a next step.
A source-aware six-module project with actual licensed microscopy, original supplied records and preserved category archives. Completion records evidence, not automatic mastery or clinical/practical authorization.
Before public feedback, state a question, competing mechanisms and a six-module plan covering microscopy, growth/metabolism, genetics, ecology, host interpretation and control. Define independent units, sources, uncertainty and discriminating predictions.
Evidence to retain: Original plan and prediction snapshot, selected source/version/reuse and exposure declaration. Prior six category plans/reports remain archived rather than overwritten.
Open both supplied actual public-domain micrographs and provenance. Record visible features, an alternative, creator/source/hash/reuse and source-label exposure; separately perform the original calibrated image calculation without assigning that scale to the photos.
Evidence to retain: Actual image-specific observations and source audit, plus separate calibration units/error bounds. No fabricated species identification, viable-cell count or physical image scale.
Reconstruct growth/metabolic, genetic and ecological supplied records, then connect host/immune and control evidence with distinct denominators and competing causal explanations. Retain code or reproducible arithmetic, changed cases and rejected inferences.
Evidence to retain: All six module capabilities represented by calculations and mechanism reasoning with source identity, observation/model distinction and no operational biological or personal clinical claims.
Submit a coherent first report linking question, evidence, uncertainty, alternatives and limitations before receiving critique. Declare public answer exposure and assistance.
Evidence to retain: Original immutable report and linked task responses; submission alone is not verified scientific reasoning.
Obtain actual critique from a human reviewer addressing the evidence and assumptions, then preserve the original and a reasoned revision log.
Evidence to retain: Reviewer identity/content and substantive rationale, original/revised report and explanation of accepted/rejected corrections. No fabricated review or approval.
Attempt a genuinely unfamiliar reviewer-supplied source or numerical packet before its solution is revealed. Explain which inferences survive changed assumptions and which need new evidence.
Evidence to retain: Unfamiliar first response, source/exposure and human review. Actual supervised laboratory competence, independent learner validation, clinical authority and institutional equivalence remain separate and may be unmet.
Retain raw-data identifiers, provenance, units, prior predictions, methods, calculations, competing explanations and safety boundaries. State explicitly when data are simulated, supplied or genuinely observed. This saved text cannot verify a real experiment; do not invent measurements or completed supervised work. Maximum 12,000 characters; retain larger evidence files separately and cite them.
A nonempty report is only a saved submission. First and revised reports remain separate and cannot be overwritten. A critique of the first report is required before retaining one revision.
A new record changes image provenance, recovery, ancestry, community function, host sampling or control specificity. Defend an inference across all six modules without reusing an exposed worked response as independent evidence.
A human must supply the actual new task, evidence and conditions before the learner responds. The transfer-direction text is not itself an unseen assessment. Use the retained assignment form after the first project; exposed public task identifier: expanded-public:college-microbiology.
Science backups preserve each course's separate responses, exposure, rubric reviews and project history. Older backups do not reset attempts or erase known exposure. Divergent same-version immutable responses and incompatible same-version project histories are rejected rather than silently replaced; preserve both files for review.
Original LearnAI explanations and tasks use the references for scientific scope. Any attributed dataset or image has its own declared provenance. Laboratory or field activity requires an appropriate educator-approved protocol; no hazardous real-world operation is required by the guided path.
Science work is stored locally in this browser, separately from mathematics and AI. It does not sync automatically. Use this science-only backup; this checkout's older all-subject export does not include science.
A legacy file is available only when the complete snapshot fits the 36 MB limit. For larger snapshots, prepare a complete file set below; never discard notebooks to make a smaller file.
Matching imported notebook drafts replace their current drafts. Omitted lessons and known first attempts remain. Export your current work first if you want to retain both drafts.
A file set can hold more work than one science file. Keep its manifest and every part together. Each file is limited to 36 MB. This format needs this version of LearnAI or a compatible newer version; older versions can use only the legacy single-file backup above.
The set includes every saved science course and global continuation at this moment. Later edits need a new export. A conflicting tab exports its own visible work, which may differ from saved work in another tab.
0 file(s) selected. File order and filenames do not determine identity. Missing, changed or conflicting parts block the entire restore; no good part is applied by itself.
Science storage: loading. No account or remote computation is required.
