BURIAN, R.M.; Virginia Tech: On the Difficult of Using Nucleotide Sequences Alone to Define Genes

There has been considerable debate recently about the status of molecular concepts of the gene. Some authorities have argued for a concept, definition, or delimitation of genes as sequences of nucleotides without including the connection of the sequence to particular functions or effects in the definition. Such characterizations of genes might be linked in some way to various other ways of defining or delimiting genes in terms of the products made from those genes or the phenotypes with which they are associated, but, unlike other concepts, these strictly molecular characterizations would yield strictly intrinsic delimitations of genes; each particular nucleotide sequence defines a gene. Thus, in What Genes Can’t Do (MIT Press, 2003) Lenny Moss argues for the importance of his “gene-D” concept, according to which genes are picked out as sequences of nucleotides without taking into account what product(s) they yield or the phenotypes they cause or affect. I examine some difficulties raised by this concept and argue that a concept that employs nucleotide sequences alone to delimit genes is of very little use. I argue that a gene concept should provide grounds for saying which nucleotide sequences count (or, better, ought to count) as genes and which sequences ought not count as genes. I also argue that attempts to provide strictly intrinsic characterizations of genes as nucleotide sequences are not likely to satisfy this criterion. On this basis I conclude that significant difficulties are raised by the use of determinate nucleotide sequences (and nothing else) to define genes. I conclude that gene identification needs to take account of (at least some of) the effects, in the cellular or organismal context, that follow from the presence of the nucleotide sequences consider as candidates for the status of being a gene.