Creating a New Approach to Archiving Human Genetic Data

New technology may provide a faster way to archive human genetic data.

A genome is a long genetic sequence written in four-letter codes--three billion of them, in the case of the human genome. But what the meaning is or how the code is deciphered remains unclear to this day. Typically, this is left to professional annotators who use information from a number of sources (for instance, knowledge about similar genes in other organisms) to determine where a gene starts, stops, and what it does. Even the "gold standard” of professional annotation is an extremely slow process. However, new technology may provide a faster solution.

The U.S. Public Library of Science (PloS) is utilizing the Internet to improve access to data, and to facilitate discussion and the understanding of science. PLoS Biology is presenting information on an independent project working towards the same goals. The researchers, which include Andrew Su, Ph.D., a senior investigator from the Computational Biology Group of the Genomics Institute of the Novartis Foundation (San Diego, CA, USA) and colleagues, described their efforts to establish a "Gene Wiki”--an online repository of information on human genes in the July 2008, issue of the journal PLoS Biology. They envision a network of articles, created by a computer program and enhanced by user comments, which will describe the relationship and functions of all human genes.

There is a lot of potential information about any given gene--its name, sequence, position on a chromosome, the protein(s) it encodes, other gene(s) with which it interacts. Presenting this information is referred to as "gene annotation.” As information may come from many different researchers working independently, it is vital that resources exist to collect the information together.

Existing annotation libraries include gene portals and model organism databases--however, the information stored in these is considered definitive, which requires continual updates by specific experts and formal presentation of data. The work reported in PLoS Biology is intended to allow a much more flexible, organic accumulation of science, with all readers also able to edit and add to the Gene Wiki pages.

To stimulate the development of this Wikipedia-based resource, Dr. Su and colleagues developed a system that automatically posts information from existing databases as "stub” articles on Wikipedia. A computer program downloads information from one system, formats it according to Wiki formatting and the "stub” template that the scientists have designed, and--if a page does not already exist for that gene--posts the data on Wikipedia.

The investigators are convinced that their stubs will germinate into more detailed posts of from scientists who encounter them on Wikipedia--and they report that, so far, they appear to be succeeding, as the absolute number of edits on mammalian gene pages has doubled.


Related Links:
Genomics Institute of the Novartis Foundation