Figure 1

Gene fusion discovery pipeline for the Ivy Center SOLiD single-end data. Reads were aligned to the hg19 assembly using bioscope-1.3 software by Life Technologies. RPKM values were calculated for each exon, followed by modified cancer outlier profile analysis (COPA). If any of the exons of a gene displayed outlier expression in a sample, then the read distribution across that gene was evaluated for that sample. If either the 3′ or 5′ end of the gene had a considerably lower RPKM value compared to the other, the gene was further evaluated for fusion. All partially mapped sequences to a potential fusion breakpoint were extracted. One or more consensus sequence was generated and translated to base space format from the color space format. The consensus sequences were then aligned to the hg19 human genome using UCSC BLAT. If part of the consensus sequence mapped to the known exon and the other part uniquely mapped to the genome, the sequence was considered a potential fusion sequence. All potential fusion sequences were validated with fusion qPCR followed by Sanger sequencing.