Architecture

A modular pipeline

Each step is an independent function. Run the full pipeline, specific ranges, or individual steps.

Step 0

PFam model retrieval

Fetches the seed HMM by PFam accession, or uses a locally supplied model file, as the starting point for the search.

Steps 1 – 5

Protein mining (raw model)

HMM search, protein filtering, domain extraction, alignment, and model adjustment with WebLogo generation.

Steps 6 – 10

Protein mining (adjusted model)

Repeats mining with the species-specific model to capture additional members missed by the raw PFam model.

Step 11

MEME motif discovery

De novo identification of conserved protein motifs across all family members.

Step 12

Gene structure

Maps exon/intron boundaries for each family member from the genome annotation.

Steps 13 – 14

Promoter & cis-regulatory analysis

Extracts upstream promoter sequences and tests them for transcription-factor binding motif enrichment against JASPAR with FIMO/AME.

Step 15

Pseudogene detection

Scans the genome for degenerate, non-functional copies of the gene family missed by protein-based mining.

Steps 16 – 18

Gene tree inference

CDS extraction, nucleotide alignment, and maximum-likelihood tree inference with IQ-TREE, with optional bootstrapping.

Step 19

Composite gene tree plot

Combines the tree, domain architecture, gene structure, MEME motifs, and cis-element hits into one publication-ready figure.

Step 20

Physicochemical properties

Molecular weight, isoelectric point, GRAVY index, and other protein properties for all family members.

Step 21

Cellular localization

Predicts subcellular localization for each protein with WoLF PSORT.

Steps 22 – 25

Synteny & duplication analysis

Identifies tandem and segmental duplications with MCScanX, computes Ka/Ks ratios, and renders a circos-style synteny plot.

Step 26

Summary of results

Compiles every step's outputs into a single, publication-ready summary report for the family.

Quick example

Run the entire pipeline with a single command.

# Full pipeline (steps 0–26) gftoolkit gfscan -c /path/to/config.cfg # Specific step range gftoolkit gfscan -c config.cfg -n 1-10 # Single step by name gftoolkit gfscan -c config.cfg -n 11
Arge L. W. P., Stupar R., Morrell P. L. (2026) GFToolkit: a reproducible, containerized, one-command framework for genome-wide identification and evolutionary characterization of gene families. bioRxiv.