Example report
What a report looks like.
A short, fictitious report showing how results, uncertainty and limits are presented. Every number on this page is synthetic.
Synthetic data — fictitious example. These are not Recanthia results, not real patients and not real genes.
1. Question
Which regulatory programmes are associated with response to candidate treatment X in a cohort of 120 synthetic tumour samples (60 responders, 60 non-responders)?
2. Cohort and quality control
- 120 synthetic samples, 2 batches; 4 excluded after quality control
- Response defined before analysis (binary, pre-registered in the analysis plan)
- Covariates: batch, age group, prior treatment
3. Main result
- R01–R03 show the largest differences, with intervals that exclude zero in this synthetic cohort.
- R07–R08 are uncertain: their intervals are wide and close to zero.
- Association is not causation: these are hypotheses to test experimentally.
4. Robustness
| Check | R01 | R02 | R03 |
|---|---|---|---|
| Resampling (1,000×) | stable | stable | stable |
| Adjusted for batch and covariates | stable | weaker | stable |
| Alternative response definition | stable | stable | weaker |
5. Limits
- Single synthetic cohort; no independent validation
- Possible residual confounding by treatment history
- Inferred activity is an indirect measure of regulation
6. Proposed next steps
- Replicate in an independent cohort with the same response definition
- Functional test of R01 in two cell models
- Refine the population estimate for the R01-high profile