oblivionis028/bioinfo-collinearity-kaks-pipeline

This repository is a reusable template for a specific type of genetics research called comparative genomics. More specifically, it automates two related analyses: collinearity analysis, which compares the physical arrangement of genes across two genomes to find regions that have stayed in the same order over evolutionary time (these are called syntenic or collinear regions), and Ka/Ks analysis, which measures whether a pair of genes has been under selective pressure during evolution. A Ka/Ks value below 1 suggests the gene has been conserved by natural selection, a value near 1 suggests neutral drift, and a value above 1 suggests the gene may have evolved new functions. The README is written in Chinese. The pipeline starts from two standard bioinformatics file types: a GFF or GFF3 file that annotates where genes sit on a genome, and a protein FASTA file containing the protein sequences. From those, a Python script prepares the input formats required by a tool called MCScanX (run through a program called TBtools), which identifies the collinear gene pairs. A second step uses a tool called DIAMOND to run an all-versus-all protein similarity search, which MCScanX needs as part of its input. Once collinear gene pairs are identified, additional Python scripts filter the results down to pairs involving specific target genes of interest, extract the coding DNA sequences (CDS) for those gene pairs, and validate the sequence lengths. The final step feeds those sequences into TBtools again to calculate Ka/Ks values for each pair. The repository contains only the pipeline scripts and example configuration files. It deliberately excludes all actual genome data, annotation files, and results, both to keep the repository small and to avoid sharing potentially sensitive research data. The .gitignore is set up to block common genomics file types from being committed by accident. Dependencies include Python 3, DIAMOND, TBtools-II, and optionally gffread.