Machine learning approaches to identify core and dispensable genes in pangenomes

Abstract

A gene in a given taxonomic group is either present in every individual (core) or absent in at least a single individual (dispensable). Previous pangenomic studies have identified certain functional differences between core and dispensable genes. However, identifying if a gene belongs to the core or dispensable portion of the genome requires the construction of a pangenome, which involves sequencing the genomes of many individuals. Here we aim to leverage the previously characterized core and dispensable gene content for two grass species [Brachypodium distachyon (L.) P. Beauv. and Oryza sativa L.] to construct a machine learning model capable of accurately classifying genes as core or dispensable using only a single annotated reference genome. Such a model may mitigate the need for pangenome construction, an expensive hurdle especially in orphan crops, which often lack the adequate genomic resources.

Calmodulin and calmodulin-like gene family in barley: Identification, characterization and expression analyses

“…Pan-genome enabled the characterization of the genetic diversity present in a species. In plants, core genes are often associated with essential metabolic processes, while dispensable genes are related to adaptive functions such as disease resistance and stress responses (Danilevicz et al, 2020;Yocca and Edger, 2022). HvCaM/CML was a large gene family, and according to the released firstgeneration barley pan-genome data (Jayakodi et al, 2020), 81 (95.3%) HvCaMs/CMLs were core genes (Figure 6).…”

Genomic, Phylogenetic and Physiological Characterization of the PAH-Degrading Strain Gordonia polyisoprenivorans 135

“…The pangenome of the three strains (Figures 3 and S3) consists of 6259 genes, with 4089 classified as core genes. Dispensable genes, defined as those absent in at least one of the strains [83], comprise 674 genes (10.8%). The G. polyisoprenivorans pangenome was analyzed using a dataset comprising three strains: G. polyisoprenivorans 135 and G. polyisoprenivorans C (CP073075.1) which are closely related, and G. polyisoprenivorans VH2 (CP003119.1), which is phylogenetically distinct from this pair.…”

“…The pangenome of the three strains (Figures 3 and S3) consists of 6259 genes, with 4089 classified as core genes. Dispensable genes, defined as those absent in at least one of the strains [83], comprise 674 genes (10.8%).…”

The Progression in Developing Genomic Resources for Crop Improvement

“…ML is widely used for crop improvement; some of the case studies include plant–pathogen interactions [ 210 ], traits, and phenotyping [ 211 ], and applications include at the molecular level in plants [ 212 ]. The use of ML in plant genomics has increased in the last decade [ 213 ]; applications include the classification of genes into active and inactive genes in maize [ 214 ], identifying genome crossovers [ 215 ], identification of near-complete genetically fixed genomic regions [ 216 ], gene regulatory networks in maize [ 217 ], gene prediction with deep learning with a variety of architectures [ 218 ], diagnosis of pests and disease [ 219 ], gene prediction concerning climatic conditions [ 220 ], predicted gene expression levels from genomic sequence data [ 221 ], identifying variants based on short-read sequence alignments [ 222 ], and classifying genes as core and dispensable genes [ 223 ].…”