Skip to content

Bioinformatics Bit 1: Sample Naming Best Practices

Updated: Jan 27, 2025

Why Good Sample Names Matter

Poor sample naming can lead to confusion, errors in analysis, and difficulties in data sharing. Well-structured sample names make your data more:

  • Readable for humans
  • Parseable for computers
  • Reproducible for future analysis
  • Shareable with collaborators

Golden Rules for Sample Names

  1. No spaces - Use underscores (_) or hyphens (-) instead
  2. No special characters - Stick to letters, numbers, and underscores/hyphens
  3. Start with letters - Don't begin names with numbers or symbols
  4. Be consistent - Use the same format across all samples
  5. Be informative - Include key metadata but remain concise
  6. Zero pad numbers - Ensure proper sorting and consistency

Zero Padding: Why It Matters

Without zero padding, computer sorting can give unexpected results: 

sample_1
sample_10
sample_2
sample_3
With zero padding, files sort naturally: 
sample_01
sample_02
sample_03
sample_10

Example Structure

[Condition]_[Replicate]_[TimePoint]
Good Examples:

WT_rep01_D00
KO_rep02_D07
treated_rep01_02h
control_rep02_24h
Bad Examples: 
Sample 1                  # Contains space
2nd-replicate             # Starts with number
RNA-seq@timepoint2        # Contains special character
RNA_seq_rep_1             # Inconsistent with other names
WT_Rep1_D0               # Missing zero padding

Pro Tips

  • Document your naming convention in your project README (or similar metadata file)
  • Keep names concise and short
  • Include critical metadata but don't overload
  • Consider future sorting when structuring names
  • Use lowercase to avoid case-sensitivity issues
  • Always zero pad numbers (01, 02... instead of 1, 2...)
  • Decide on padding width before starting (e.g., 01-99 needs two digits)

Need help with sample names? Contact the CGDS Core -- CGDS@mdibl.org