GET_HOMOLOGUES

GET_HOMOLOGUES performs pan-genome clustering and comparative analyses of homologous gene families and, with the GET_HOMOLOGUES-EST extension, analyzes pantranscriptomes from RNA-seq and transcriptomic data to classify core and accessory sequences and estimate pangenome properties.

Key Features:

• Homology clustering algorithms: Supports clustering with bidirectional best-hit, COGtriangles, and OrthoMCL algorithms.
• Domain composition scanning: Integrates HMMER3-based domain composition screening to inform cluster stringency.
• Alignment and synteny filters: Allows pairwise alignment coverage cutoffs and selection of syntenic genes to refine homologous clusters.
• Consensus clustering: Produces robust homologous gene families via consensus clustering from combinations of algorithms and filters.
• Pangenome and core genome construction: Constructs, interrogates, and generates graphical representations of core genome and pangenome sets.
• Statistical pangenome models: Uses exponential and binomial mixture models to estimate theoretical pangenome and core genome sizes.
• Pangenome trees and comparative analyses: Computes pangenome trees and performs comparative genomics tasks including identification of lineage-specific genes and gene family expansions.
• Pantranscriptome support (GET_HOMOLOGUES-EST): Leverages transcriptomic and RNA-seq data to sample genic repertoires for species with large or highly heterozygous genomes.
• Presence–absence and ANI matrices: Computes presence–absence pangenome matrices and compiles Average Nucleotide Identity (ANI) matrices summarizing intra-species variation.
• Pan-genome growth simulation and private gene detection: Simulates pangenome growth from sequence clusters and identifies private genes present only in some genotypes.
• Parallelization: Parallelizes computationally intensive tasks for multiprocessor workstations and computer clusters.
• Empirical benchmarking: Has been benchmarked on datasets including 50 Streptococcus genomes (OMA annotations), 19 Arabidopsis thaliana ecotypes, and transcripts from 16 Hordeum vulgare genotypes.
• Accessory locus characterization: Enables analysis showing accessory loci tend to have lower expression, higher non-synonymous substitution rates, and enrichment for transposon components, disease resistance genes, and presence/absence-associated protein domains.

Scientific Applications:

• Microbial pan-genomics: Comparative analysis and reconstruction of core and accessory gene sets across microbial genomes.
• Plant pantranscriptomics and pangene analysis: Sampling and classification of genic repertoires in species with large or heterozygous genomes using RNA-seq-derived transcripts.
• Pangenome size estimation: Estimation of theoretical pangenome and core genome sizes using statistical mixture models.
• Phylogenetics and population structure: Reconstruction of pangenome trees and summarization of intra-species variation via ANI matrices.
• Presence/absence variation studies: Detection and characterization of accessory sequences, private genes, and loci associated with transposons and disease resistance.
• Comparative genomics: Identification of lineage-specific genes, gene family expansions, and differences in expression and substitution patterns between core and accessory genes.

Methodology:

Clusters homologous genes using bidirectional best-hit, COGtriangles, and OrthoMCL; applies HMMER3 domain scans, pairwise alignment coverage cutoffs, and synteny filters; produces consensus clusters; fits exponential and binomial mixture models for pangenome size estimation; computes pangenome trees, presence–absence matrices, and ANI matrices; uses transcriptomic/RNA-seq data for GET_HOMOLOGUES-EST benchmarking on specified datasets and parallelizes tasks for multiprocessor execution.