Data Preprocessing
Use these pages before creating a SequenceData object or before running a workflow that expects clean IDs, time columns, and sequence states.
Choose a Task
| Task | Start with |
|---|---|
| Convert sequence data between wide and long formats | Wide and Long Format Conversion |
| Check missing values before analysis | Check Missing Values |
| Create stable IDs when the source data do not have them | Assign Unique IDs |
| Standardize time-column names | Clean Time Columns |
| Replace numeric cluster IDs with readable labels | Replace Cluster ID by Labels |
Recommended Order
For most projects, convert the data to wide format if needed, clean the time-column names, make sure IDs are stable, check missing values, and then create SequenceData. Replacing cluster IDs with labels comes later, after clustering.