Toolbox for Mobile-Element Insertion Detection on Cancer Genomes

Mobile elements constitute greater than 45% of the human genome as a result of repeated insertion events during human genome evolution. Although most of mobile elements are fixed within the human population, some elements (including ALU, long interspersed elements (LINE) 1 (L1), and SVA) are still actively duplicating and may result in life-threatening human diseases such as cancer, motivating the need for accurate mobile-element insertion (MEI) detection tools. We developed a software package, TANGRAM, for MEI detection in next-generation sequencing data, currently serving as the primary MEI detection tool in the 1000 Genomes Project. TANGRAM takes advantage of valuable mapping information provided by our own MOSAIK mapper, and until recently required MOSAIK mappings as its input. In this study, we report a new feature that enables TANGRAM to be used on alignments generated by any mainstream short-read mapper, making it accessible for many genomic users. To demonstrate its utility for cancer genome analysis, we have applied TANGRAM to the TCGA (The Cancer Genome Atlas) mutation calling benchmark 4 dataset. TANGRAM is fast, accurate, easy to use, and open source on https://github.com/jiantao/Tangram .