Tutorial: Custom GMT Libraries

Canonical layout

<SET_NAME>\t<DESCRIPTION_OR_URL>\t<GENE_1>\t<GENE_2>\t...\t<GENE_N>

• SET_NAME: short identifier for the set (no tabs). Prefer A-Z, 0-9, _, -.
• DESCRIPTION_OR_URL: free text, a reference, or a link. Can be empty (- or blank after the tab).
• GENE_i: one identifier per column (e.g., HGNC symbols). All sets in a file should use the same ID system and species.

Minimal working example

TNF_SIGNALING\tKEGG hsa04668\tTNF\tTNFRSF1A\tTRADD\tTRAF2\tMAP3K7\tIKBKB\tNFKBIA\tRELA
OXPHOS_CORE\tReactome R-HSA-1428517\tNDUFS1\tNDUFA9\tUQCRC1\tCOX4I1\tATP5F1A

Common pitfalls & how to avoid them

• Mixed identifier namespaces: Only use the official gene symbols (for example from HGNC or MGI).
• Whitespace separators: tabs only between fields. Do not use spaces or commas to separate fields.
• Empty or duplicate genes: drop NA/blank entries and de-duplicate within each set.
• Oversized / undersized sets: extremely large (>5000 genes) or tiny (<3 genes) sets can distort statistics.
• Ambiguous set names: avoid tabs, newlines, or excessive punctuation in SET_NAME. Keep names unique.
• Inconsistent casing: if you choose HGNC symbols, keep them uppercase (e.g., TP53), not mixed case.

Preparing a custom library

1. Choose an identifier scheme (e.g., HGNC for human). Good practice is to keep every term in your library using the same scheme.
2. Build gene set membership: Create a set of genes for each term.
3. Write the GMT with one term per line, tabs between fields, no trailing tabs.
4. Validate with the checklist below.

Validation checklist

• All lines have ≥3 columns (name, description, ≥1 gene). Description can be empty but the tab must be present.
• No duplicate SET_NAME values.
• No duplicate gene entries within a set.
• No tabs inside gene symbols or set names.
• Consistent species and gene symbol namespace.

grep -n "  " your_library.gmt   # flags space runs; fix to tabs

Adding custom libraries to pathXcite

You can add custom libraries either via the UI or by placing files in the data folder.

1. Open Settings (1) → Gene Set Libraries (2).
2. Scroll to Custom Libraries (3) (bottom panel). Click the arrow to expand if not visible.
3. Click Add .gmt and select your .gmt file.
4. Confirm. The table lists name, term count, and file size as well as whether the file has a valid format.
5. Click Remove Selected (5) to remove a custom library.
6. Click Open Folder (6) to see where the files are stored.

Using your custom library in analyses

1. Open the Enrichment (1) module.
2. Click on the library selector (2).
3. In the library selector, you should now find the custom GMT (3).

FAQ & troubleshooting

• My library loads but shows zero terms. Check for spaces instead of tabs; ensure each line has at least three fields.
• Some genes look missing. Verify your identifier scheme matches your input list (e.g., HGNC vs Ensembl). Convert if needed.
• Duplicate set names. Ensure unique SET_NAME per row; rename with a suffix (e.g., _v2).
• Very large sets dominate results. Consider filtering sets to a reasonable size range (e.g., 5-2000 genes) before analysis.