Publication in Nature: GSF scientists decode the chicken genome

As members of an international research consortium, a group of GSF scientists led Randolph B Caldwell and Jean-Marie Buerstedde contributed to the annotation of the complete chicken genome. This first genome sequence of a bird is not only of great importance for research projects with chickens, but it will also lead to a better understanding of the previously decoded human genome.

Originally, the GSF research project aimed at identifying active genes in a particularly interesting chicken B cell line DT40. However, when the US Department of Agriculture approved funding for the sequencing of the chicken genome two years ago, an international consortium was rapidly formed by scientists from the USA, Europe and China to coordinate the research activities. While Washington University at St. Louis in the USA sequenced the genome of the Red Jungle Fowl, the ancestor of today’s domestic chicken, the GSF team worked closely with two English groups from Hinxton and Manchester on the definition of the chicken gene sequences. Genes constitute only a small proportion of the genome of vertebrates, but they are extremely important since they determine the amino acid sequences of all proteins and thus the structure and regulatory mechanisms of cells and organs. In the first step, genes in a newly sequenced genome are provisionally identified with the help of computer assisted search programs. In order to test the reliability of the computer predictions, the team from the Institute of Molecular Radiation Biology decided to sequence over new 2000 genes – approximately 10% of the total estimated chicken genes. From the start, the project’s focus was on quality not quantity. Caldwell succeeded in training and motivating 30 work students from the Munich universities to sequence and assemble the gene sequences and then check them by comparing them to known genes from other animals. In addition, thousands of short gene tags were identified in order to verify suspected genes. These investigations showed that the computer generated gene predictions have many inaccuracies, especially at the start and end regions of the genes, which still have to be corrected by follow-up projects.