The GLAD algorithm provides very good results as shown in Tab 1 for both the detection of breakpoints step and the region assignment step. Our algorithm performs better than other methods on a public reference data set (Snijders et al., 2001).
Data type | Breakpoints detected | Status correctly assigned |
simulated data | 97% | 98.9% |
karyotyping results (Snijders et al., 2001) | 100% | 99.8% |
manually analyzed profiles | 94% | status are unavailable |