Ingredient patterns
The same lag analysis as Patterns, but on the individual ingredients extracted from your meals rather than tags. Showing the top 10 ingredients whose presence before a symptom most exceeds their baseline. The number is the raw rate; the shading is how far that sits above the ingredient's baseline — so something you eat daily stays pale, and only genuinely elevated cells light up.
68 meals
14 symptoms
18 days logged
| Ingredient | Baseline | 0-6h | 6-12h | 12-24h | 24-48h | 48-72h |
|---|---|---|---|---|---|---|
| garlic powder | 33% | · | · | 64% | 36% | · |
| onion powder | 33% | · | · | 64% | 36% | · |
| baking powder | 22% | · | · | 50% | 21% | · |
| wheat flour | 44% | · | · | 71% | 36% | 14% |
| butter | 11% | · | · | 29% | 36% | 14% |
| cilantro | 28% | · | · | 50% | 14% | 21% |
| gellan gum | 28% | 21% | · | · | 50% | 21% |
| oat milk | 28% | 14% | · | 50% | 14% | 7% |
| whole wheat flour | 28% | 36% | · | 43% | 50% | 50% |
| carrot | 22% | · | · | 43% | 7% | 14% |
Exploratory only — this surfaces candidates to test by elimination, it doesn't prove cause. Ingredient signals are noisier than tags (more items, smaller counts each); trust them more after a few weeks of consistent logging.