Treat an experimental workflow diagram as a methods summary. Write the real steps, their relationships, and the final outputs before choosing icons and colors. The existing microfluidic case below demonstrates the layout method without prescribing experimental parameters.
1. List the steps before choosing the node count
Extract steps from your methods description and give each node one main action. When there are many operations, group them into stages instead of forcing every action into one long chain.
The microfluidic case shows sample injection, channel mixing, a cell culture chamber, imaging, acquisition, and analysis. Replace these with your own workflow and adjust the node count.
- Give each node a short title.
- State the input and output for each stage.
- Keep only the key stages a reader needs to understand the method.
2. Use arrows for real relationships
Arrows should express sequence, branching, merging, or feedback. Parallel tests should be drawn as branches rather than as a sequence chosen only because the layout is horizontal. For a chip diagram, distinguish fluid direction from workflow direction.
- Give every branch a clear start, condition, and output.
- Use the same layout scale for controls and treatments.
- Use a separate path and label for feedback relationships.
3. Separate experimental operations from analysis
Keep instrument operations, data acquisition, and data analysis in adjacent but distinct areas. An empty result window is more appropriate for a draft than an invented data chart.
After specifying a white background, consistent icon style, and step numbers, choose a landscape or grouped layout so nodes do not each adopt a different style.
Draw a workflow diagram for 【experiment name】. Arrange nodes in 【real step list】 and show 【serial, parallel, merge, or feedback relationships】. Divide the figure into 【experimental operations, data acquisition, and analysis output】 areas, with short English labels and consistent icons. Use a white background and landscape 16:9. Leave the results area as an empty placeholder; do not generate measurements or data curves.
4. Compare every item with the methods text
Start at the first input and follow every arrow to its output. Delete or correct any node that cannot be explained by the methods text. Then check equipment names, platform abbreviations, and English labels.
- Is the node order accurate?
- Were parallel stages incorrectly connected in sequence?
- Are chip flow arrows mixed with operation workflow arrows?
- Do readings in the figure come from your own measurements?