EthoSource: Storing, Sharing, and Combining Behavioral Data
Emı́lia P. Martins · BioScience · 2004
To date, the phenomenal growth of bioinformatics and computational biology has largely focused on the organization, storage, and analysis of sequences of As, Cs, Gs, and Ts, corresponding to nucleotides of DNA. However, most forms of biological data are far less easily manipulated. Consider, for example, birdsong. A song might be stored as an audio file, a graph describing the spectral properties of the sound, or a table of measurements describing the average qualities of the song. But birdsong may be influenced by a multitude of factors, including the genetic makeup of the singer, its past experiences, its testosterone levels, the presence of other birds, and the physical properties of nearby vegetation. Full use of measurements of these and other factors may require combination of very different types of data in entirely novel ways. Designing a database that organizes information about birdsong is therefore much harder than designing a database for DNA sequences. The future of bioinformatics lies in developing software to manipulate the diverse data generally needed to describe biological traits.