Sample Sheet guide

1 · Upload your metadata

Screening thresholds

Q1 comparison	Association measure	Priority
Categorical–categorical	Cramér’s V	≥ 0.40
Categorical–continuous	Eta	≥ 0.40
Continuous–continuous	Absolute Spearman correlation	≥ 0.40

If no variable meets the main threshold, associations from 0.20 to below 0.40 are shown. These are teaching priorities, not proof of confounding. SMD and Duncan do not select variables at Q1.

After Q2/Q3 flag a potential confounder, examine shared categories and distributions/ranges. SMD and Duncan support imbalance assessment. High imbalance or correlation alone does not establish complete confounding.

Q1 is descriptive: related specimens remain in the screen, so subjects contributing more specimens may influence it more. Technical repeats are counted once using specimen ID where available. When “Same specimen” is selected without specimen IDs, the subject ID is used.

Load a worked case study:

Upload CSV or TSVOr paste a table, including its header

No sheet loaded.

2 · Map the key columns

Sample / specimen ID
the row-level measurement identifier (e.g. SampleID, RNA_ID). Can repeat if one specimen was processed twice. Leave blank if not separately recorded.Subject / donor ID
the individual being studied (e.g. PatientID, MouseID, DonorID). Multiple rows share this ID when one subject has several specimens.Phenotype / outcome
the column defining your comparison groups (e.g. Disease_status, Treatment, Age_group).

Starting the metadata activity

Omics experimental design calculator

1 · Upload your metadata

Screening thresholds

2 · Map the key columns

3 · Choose variables to screen

4 · How are your samples related?

5 · Assess possible variable roles

6 · Review your study summary

Starting the metadata activity

1 · Upload your metadata

Screening thresholds

2 · Map the key columns

3 · Choose variables to screeni

4 · How are your samples related?

5 · Assess possible variable roles

6 · Review your study summary

3 · Choose variables to screen