GoldMiner: Gene Origin, Duplication, and Loss Analysis Tool

GoldMiner is a tool for analyzing and comparing homology gene clusters (HOC) across different genomes. It provides a complete analysis pipeline from gene cluster identification and multi-genome comparison to evolutionary event inference.

Features

• TDGFinder: Identify clusters in each genome (N = 1)
• PairLink: Connect clusters between pairwise genomes (N = 2)
• MultiCluster: Build clusters colinear network in all genomes (N ≥ 3)
• OdlRecon: Infer cluster origin, loss, and duplication events

Requirements

• R ≥ 4.0
• R packages: igraph, reshape2, dplyr, parallel, data.table
• Python 3.x
• MCL tool (for clustering analysis)

Usage

Basic Command Format

goldminer <command> [options]

Available Subcommands

[ pipeline ]
     TDGFinder      Step2: Identify clusters in each genome (N = 1)
     PairLink       Step3: Connect clusters between pairwise genomes (N = 2)
     MultiCluster   Step4: Build clusters colinear network in all genomes (N ≥ 3)
     OdlRecon       Step5&6: Clusters origin, loss and duplication inference

[ tool ]
     HocAliPlot     Dotplot of HoC alignment between two genomes

Complete Analysis Pipeline

Step 1: Genome Comparison Analysis

This tool relies on genetribe for core gene comparison between genomes. genetribe serves as a foundational framework for gene collinearity and orthology analysis.

For more details about genetribe and its functionalities, visit genetribe Website and TGT database.

genetribe core -l rice -f rice
genetribe core -l aet -f aet
genetribe core -l aet -f rice

Step 2: Identify Gene Clusters

Use TDGFinder to identify gene clusters in each genome:

goldminer TDGFinder -f data_path -b genome_bed_path -o output_path -s subgenome_info_file -t num_threads -d max_distance

Parameters:

• -f/--one2many: one2many file path
• -b/--bed: bed file path
• -o/--output: output file path
• -t/--threads: threads to use in parallel
• -d/--distance: the max distance of tandem genes
• -s/--subgenome: the file of subgenome

Step 3: Connect Clusters Between Pairwise Genomes

Use PairLink to connect clusters between pairwise genomes:

goldminer PairLink -l first_genome_prefix -f second_genome_prefix -o output_path -c cluster_path -i iblocks_path -b blocks_path -t num_threads

Parameters:

• -l/--first: Prefix name of first file
• -f/--second: Prefix name of second file
• -t/--threads: threads to use in parallel
• -o/--out: link file output path
• -c/--clu: cluster file path
• -i/--iblocks: iblocks file path
• -b/--blocks: blocks file path

Step 4: Build Clusters Colinear Network

Use MultiCluster to build clusters colinear network in all genomes:

goldminer MultiCluster -d output_directory -i input_directory -t genome_list_file -p output_prefix -c cluster_files_directory

Parameters:

• -d/--dir: directory of output files
• -i/--input: directory of input files
• -c/--cluster: directory of cluster files
• -t/--table: genome list file
• -p/--prefix: prefix of output file

Step 5&6: Infer Evolutionary Events

Use OdlRecon to infer cluster origin, loss, and duplication events:

goldminer OdlRecon data_path file_prefix selected_groups

Parameters:

• path: Path to the data files
• prefix: Prefix for the data files
• selected_groups: Comma-separated string of selected groups (e.g., 'D,A,B,Thinopyrum,Secale,Hordeum,Avena,Brachypodium,Oryza,Zea')

Output Files

• .cludb files: Gene cluster database files
• .clu files: Gene cluster information files
• .link files: Mapping files between genome clusters
• .matrix files: Wide-format matrices for gene cluster comparison

Citation

If you use GoldMiner in your research, please cite:

License

MIT License

Contact

Author: Xiaoming Xie
Email: xiexm@cau.edu.cn