Gradient Biotech

Overview

Infectious Disease workflows link pathogen, host, exposure, and intervention evidence to clinical outcomes and person-place-time epidemiology in one episode-centered study workspace. Each use case below maps a common research question to the platform capabilities, data requirements, and suggested analysis path.

Use case index

Use caseResearch questionKey capabilities
Infection timelineHow do exposure, symptoms, specimens, treatment, and outcomes align for each episode?Longitudinal episode timeline, missingness reporting, index and risk windows
Person-place-time epidemiologyWhat are the rates, epidemic curve, and stratified patterns in this cohort?Attack rate, incidence, hospitalization, case-fatality, denominators, stratification
Cohort outcomesDo outcomes differ between groups, and how does time-to-event compare?Cohort comparison (RR/OR), chi-square/Fisher, Kaplan-Meier, log-rank
Outbreak investigationWhich episodes are plausibly linked, and do they form candidate clusters?Exposure/contact network, threshold sensitivity, genomic-relatedness overlay, cluster investigation, link- and cluster-level adjudication, lineage dynamics, facility co-location
Intervention effectivenessDo outcomes differ by treatment or vaccination, once time bias and confounding are addressed?Landmark treatment outcome, treatment × resistance, observational VE, breakthrough
Reproduction numberIs transmission intensity rising or falling, and how fast is the epidemic growing?Cori instantaneous Rt with credible intervals, epidemic-growth rate, doubling time
Epidemic surveillance and forecastingHow is a public-health signal trending across jurisdictions, what's the near-term forecast, and is a leading indicator giving early warning?Connected surveillance sources, epidemic curve + Rt, cross-geography/signal comparison, multi-model forecast + backtest, early warning, suppressed maps
Host-pathogen integrationHow do outcomes vary by pathogen lineage or host feature from other areas?Cross-area linked results, feature-grouped outcomes, preserved provenance
Cross-area translationalHow do microbial, immune, and host results connect to infection episodes and outcomes?Shared subject/episode identity, versioned result linking, grounded interpretation

Who these use cases serve

AudienceTypical goals
Infectious-disease researchersLink pathogen, host, intervention, and outcome evidence in reproducible cohorts
Hospital epidemiology researchersEpisode timelines, ward context, and intervention evaluation
Public-health research teamsPerson-place-time analysis and uncertainty-aware reporting
Vaccine and anti-infective teamsBreakthrough and treatment-response cohorts with clinical endpoints

Common data requirements

Most use cases start with an infectious-disease study whose central entity is the infection episode:

  • Participants and episodes with index dates, case status, and risk windows
  • Outcomes with severity, event_observed, and time_to_event_days
  • Denominators (population at risk, person-time) for population rates
  • Linked results referencing versioned Microbiology/CompBio/Immunology artifacts by episode

Participants, episodes, exposures, interventions, outcomes, linked results, and denominators can each be bulk-loaded via CSV import, rather than entered one at a time.

Choosing a starting point

If your primary data is…Start with…
Episodes with dated eventsInfection timeline
Cases plus population denominatorsPerson-place-time epidemiology
Two groups with outcomesCohort outcomes
Exposures, contacts, and isolate relatednessOutbreak investigation
Treatments or vaccination status with outcomesIntervention effectiveness
Dated episodes over an epidemic periodReproduction number
Public surveillance feeds across jurisdictionsEpidemic surveillance and forecasting
Episodes plus linked pathogen/host resultsHost-pathogen integration
Multiple areas sharing subjectsCross-area translational

What these use cases are not

These workflows support exploratory and translational research. They are not operational contact-tracing, mandatory-reporting, diagnostic, or prescribing systems, and they never assert transmission, causation, diagnosis, susceptibility, or outbreak status from association alone. All AI interpretation outputs include research disclaimers and cite computed metrics only.