VaraHelp Center
Contents — Tap to expand

Parameters · Lanes · Reproduce

Orchestration and computation once an experiment is running: parameter formulas, the lane flow view, and one-click reproduce into a new version.

① Parameter panel

Experiment top [Parameters (N)] button opens:

  1. Add: name + value + unit + (optional) formula + (optional) notes
  2. Edit inline, auto-saves
  3. ⋮ menu to delete / reorder

② Formula chaining

Parameters can hold formulas that chain off other parameters. Change A and B recomputes automatically. Example:

WellsCount = 12
PerWellVol = 1000 μL
TotalVol   = WellsCount × PerWellVol    ← formula
            (auto-recompute when WellsCount changes)

③ Auto unit conversion

  1. μL ≥ 1000 → shown as mL (12000 μL = 12 mL)
  2. mL ≥ 1000 → shown as L
  3. mg ≥ 1000 → shown as g
  4. Reverse: 0.5 mL → 500 μL auto

④ Lanes: list and flow view

The experiment page groups sub-experiments into lanes based on their dependencies. The main line is the longest chain from start to finish, the critical path: if any step on it is delayed, the whole experiment is delayed. Other steps that can run at the same time get a branch lane for each connected chain. Three or more steps with exactly the same steps before and after them (for example several samples prepared separately) are merged into one "N in parallel" lane. The list and the flow view use the same lanes and numbers; when there is only one line, no lane titles are shown.

The list is the default. Each row shows the step count, source, dependencies, versions and runs, and the [+] next to a lane title adds a sub-experiment at the end of that line (with a single line, the button is at the end of the list). Before any order is set, you can change the sequence with the reorder button in the toolbar or Move up / Move down in the ⋯ menu; once order is set, positions follow the dependencies, so use "Set order" instead. Use the switch above the list to open the Flow view (desktop only); your choice is remembered.

In the flow view, lane names are on the left and each card is a sub-experiment. Click a card to see what has to finish before it (blue) and what it affects (green). The small panel at the top right lets you jump to those sub-experiments, open the step, or set its order. Click the card again, click empty space or press Esc to clear, and use ← → to walk along the lines. The Critical pathtoggle highlights the main line; the search box finds a sub-experiment by name or number and Enter jumps to it; click or drag the overview at the bottom to jump anywhere in the graph. Zoom goes from 85% to 130%, and "Fit width" scales the graph to the window width.

To switch a sub-experiment's version, use the step detail page.

Order between sub-experiments: when you import a paper or paste a method, the AI also marks which sub-experiments must wait for others. On the review page each card shows "After #2, #3" with an "AI inferred" tag, and you can click "Edit" to correct it. On the experiment page, click "After #N" in the list, or choose "Set order" from the ⋯ menu on the row, then check which sub-experiments this one has to wait for. Options that would create a loop (its own downstream) cannot be selected.

⑤ One-click reproduce

Experiment top [Reproduce] → creates a new version (v4 = v3 fork) inheriting all steps + params. Modify any step, compare with the parent.

Faster path: tell Vara "v3's CD86 100ng is too low, try 50ng" → AI outputs 2 action cards ([New v4 iteration] + [Change CD86 = 50ng]), click confirm to auto-complete.

⑥ Pilot ↔ Formal merge

Do Pilot first (small-scale trial) → once it works, [Merge to formal] copies the Pilot data into the Formalexperiment. Pilot doesn't pollute main data.