The term 'phene' was evidently coined as an obvious parallel construct to 'gene'. Phene is to Phenotype as Gene is to Genotype, and Similarly Phene is to Phenome as Gene is to Genome. However the term is not commonly used by geneticists nor taught in genetics courses.
More specifically, a Phene is an abstract concept describing a particular characteristic which can be possessed by an organism. Whereas Phenotype refers to a collection of Phenes possessed by a particular organism, and Phenome refers to the entire set of Phenes that exist within an organism or species.
It is important to note that the word phenotype was originally used to refer to both the trait/character itself (e.g. the blue eyes phenotype) and the set of traits/characteristics possessed by the organism (clair's eye-colour phenotype is blue). Whilst this definition is still used in many places, the lack of distinction can make indepth explanations confusing and thus use of the term Phene becomes necessary.
Due to the distinct chemical and physical properties of the nucleotides in the DNA and some mutations being "silent" (that is, not altering gene expression), the DNA primary sequence may also be a phene. For example, A-T and C-G base pairs are differently resistant to heat (see also DNA-DNA hybridization). In a thermophilic microorganism, "silent" mutations may have an effect on DNA stability and thus survival. While being subject to evolution, natural selection affects the primary sequence directly in this case, with or without it being expressed.
Consider, for example, a mutation that makes a zygote abort development as a young embryo. This mutation, obviously, will not spread, as it is quickly fatal. It is not the mutated nucleotide that is selected against, but the fact that due to this mutation, the phene (a key enzyme or developmental factor for example) does not get expressed.
Compare a (fictional) kind of mutation that breaks the DNA strand in a crucial position and defies all attempts to repair it, leading to cell death. Here, the mutated and unmutated DNA sequences would be phenes themselves; it is the changed primary sequence itself which by failing would cause death, not the corresponding polypeptide.
See also Dawkin's concept of extended phenotype.
According to Anatoly Ruvinsky, The University of New England, Australia, the term was introduced by the influential Russian geneticist Alexander Serebrovsky in the 1920s. This assertion would seem to be supported by this article Whilst it is possible that the term has been coined independently several times, the basic definition of "phene is to phenotype as gene is to genotype" appears to be shared, though there is some disagreement over precisely what this phrase means.
'Phene' is not commonly used by geneticists nor taught in most genetics courses. It has a technical usage in the OMIA (Online Mendelian Inheritance in Animals) database.