
thumb|368x368px|Pangenome analysis of Streptococcus agalactiae genomes made with Anvi'o software whose development is led by [[A. Murat Eren. Genomes obtained from Tettelin et al. (2005). Each circle corresponds to one genome and each radius represents a gene family. At the bottom and at right are localized the core genome families. Some families in the core may have more than one homologous gene per genome. In the middle, at the left of the figure the shell genome is observed. At the top left are shown families from the dispensable genome and singletons. ]]
thumb|368x368px|Pangenome analysis of Streptococcus agalactiae genomes made with Anvi'o software whose development is led by [[A. Murat Eren. Genomes obtained from Tettelin et al. (2005). Each circle corresponds to one genome and each radius represents a gene family. At the bottom and at right are localized the core genome families. Some families in the core may have more than one homologous gene per genome. In the middle, at the left of the figure the shell genome is observed. At the top left are shown families from the dispensable genome and singletons. ]]
In the fields of molecular biology and genetics, a pan-genome (pangenome or supragenome) is the entire set of genes from all strains within a clade. More generally, it is the union of all the genomes of a clade. The pan-genome can be broken down into a "core pangenome" that contains genes present in all individuals, a "shell pangenome" that contains genes present in two or more strains, and a "cloud pangenome" that contains genes only found in a single strain. Some authors also refer to the cloud genome as "accessory genome" containing 'dispensable' genes present in a subset of the strains and strain-specific genes. Note that the use of the term 'dispensable' has been questioned, at least in plant genomes, as accessory genes play "an important role in genome evolution and in the complex interplay between the genome and the environment". The field of study of pangenomes is called pangenomics.
Discovered by embedding cosine similarity (sentence-transformers MiniLM, 384-dim).