Specificity values Orotic acid (6-Carboxyuracil) typically fall close to 1, such that the denominator for LR+ is usually much smaller than the denominator for LR2. As a result, the values for LR+ are usually much larger than those for LR2. An LR+ means the codes are moderately good for detecting HF, means the codes are very good. Three other validation statistics of interest were PPV, NPV, and kappa score. NPV was equal to the number of true negatives divided by the total number of cases not coded for HF. Where available, we abstracted statistics for definite and possible cases of HF, though the number of categories reported depended on the choice of reference standard. We believe this is explained by an intron sequence enrichment that occurs as a result of formalin fixation of RNA. We note that in another study using FFPE tissues more than 50% of the reads are intronic. We have excluded the possibility of genomic DNA contamination in our FFPE RNA preparations by use of criteria: DNAase I treatment, and confirmation by TaqMan assays for gDNA. The increased proportion of intronic reads from FFPE specimens may reflect selective degradation of cytoplasmic RNA by RNase during formalin fixation. This study demonstrates the technical feasibility and potential biomedical value of being able to detect fusion transcripts in archival tumor specimens having attached clinical records. Although the average frequency of detected fusion transcripts is relatively low per patient, plausibly attributable to the low quality of FFPE RNA-Seq libraries, the frequency of fusion events found in our cohort nevertheless appears to have prognostic significance. Many of the identified fusion partner genes belong to the kinase, phosphatase and ubiquitin ligase families, which are attractive pharmaceutical targets in oncology. Both fusion frequency and tumor prognosis may be linked to cancer genome instability, which can Chrysophanol-8-O-beta-D-glucopyranoside generate chromosome rearrangements and fusion transcripts. In conclusion, this study significantly enriches the current understanding of breast tumor complexity by discovering a large number of novel fusion transcripts. It confirms one of the challenges of cancer therapeutics, namely that each cancer is different and personalized treatment is needed.