Common Integration Sites (CISs) are the transposon insertion sites observed in small windows of genome by two or more tumors. The window sizes are determined by considering the likelihood of observing such occurrences by chance.
We used following window sizes for approximately 1,000 insertion spots to limit false discovery rate:
To determine the window sizes, we simulated random transposon insertions at TA sites in approximately 2Gb mouse genome by computer based on two characteristics of SB transposon (Vigdal et al., 2002: J Mol Biol.):
This procedure is similar to that of the retrovirus random integration model (Suzuki et al., 2002: Nat. Genet.).
However, there is a problem in this random SB transposition model, a local hopping. In germline cells, 50-80% of SB transpositions are located within 10-25 megabases of the donor site (Horie et al., 2001: Proc Natl Acad Sci U S A.). In somatic cells, the percentages of location transpositions are reduced, but researchers have to adjust the window sizes in donor site chromosomes.
Candidate genes are usually the nearest neighboring RefSeq gene from each integration sites. However, we do not apply this rule strictly toward common integration sites. We choose CIS candidate genes based on directions of insertions and gene functions. In addition, we may have missed the latest additions and modifications of gene definitions.
If you notice a mistake, please contact to our help desk by e-mail.
Please visit UCSC Genome Bioinformatics Homepage (http://genome.ucsc.edu). There is a step-by-step instruction how to create user custom tracks.