The Alignathon is a collaborative project to assess whole genome aligners and promote development of the field of whole genome alignment. Alignathon is insprired by the Assemblathon project, an iterative collaborative competition to assess the state of the art in de novo genome assembly.
As more and more genomes are sequenced to fill out the phylogeny of vertebrates in small independent projects and in larger coordinated efforts like Genome 10K, there is going to be a strong desire to know the evolutionary relationships between the genomes. There is coming a time quickly when whole genome alignment is going to be very much in demand.
With that near future in mind we are organizing a collaborative project to assess whole genome aligners. We have designed a test suite comprised of four problems. We present here three sets of simulated genomes created using the Evolver genome evolution simulation suite (controlled by evolverSimControl) and one real data set comprised of the 20 fly genomes.
The combination of using the most advanced genome evolution simulator available along with a set of real genomes provides a test suite of similar sized problems for aligners and a way to compare metrics that don't require a known truth to those that do.
The Evolver simulator has a gene model, models both inter and intra chromosomal evolution and keeps track of an evolving mobile element library that stochastically adds content to the genome under simulation. All of this information is available to the final analysis and will allow us to determine how well aligners perform in gene regions versus neutral regions, near inversions, in regions of recent genomic rearrangement etc.
Good luck, and thanks for being a part of the project!