Cancer genomes are marked by genomic instability and massive rearrangements. Recently, many exotic mechanisms have been proposed as mechanistic explanations for these rearrangements. For example, the breakage-fusion-bridge (BFB) mechanism, proposed over seven decades ago, has seen renewed interest as a source of genomic variability and geneamplification in cancer. Here, we formally model and analyze the BFB mechanism, the first rigorous formulation of the mechanism. Using this model, we show that BFB can achieve a surprisingly broad range of amplification patterns, and describe efficient combinatorial algorithms to characterize patterns consistent with BFB. An extensive analysis of simulated, cell-line, and primary tumor data reveals the existence of BFB. Our results also suggest that BFB may be hard to detect under heterogeneity and polyploidy. Time remaining, we will also discuss other sources of variation (joint work with Shay Zakov, and Marcus Kinsella).