Both Tier-2 and Tier-3 were selected based on the expression profiling described below; Tier-2 consists of fusions with a single non-redundant read across the fusion junction and Tier-3 represents predicted Rebaudioside-A recurrent fusions with no read across the putative fusion junction. We present here novel evidence that increasing frequency of fusion transcripts is associated with poor prognosis. This study also adds to the molecular knowledge of breast cancer complexity by identifying 118 candidate fusion transcripts and many TaqMan supported fusion transcripts, all of which are novel except TFG-. GPR128. Moreover, these fusions could be detected in single-end RNA-Seq data from aged FFPE tumor tissues by applying g Fuse, a cohort-based bioinformatics method. Among the total 118 candidate fusion transcripts identified, 3 unique fused gene pairs were recurrent and supported by TaqMan in the two cohorts of 212 total patients. The rate at which recurrent fusions were observed and the general novelty of the observed fusion transcripts in this study are in line with the previous publications about the very low recurrence of fusions in solid tumors such as 2�C7% EML4-ALK in non-small lung cancer patients. It is notable that the recent TCGA consortium efforts with large patient cohorts and fresh frozen samples assisted with whole genome sequencing identified primarily private fusion transcripts. In 416 clear cell renal carcinoma patients, 70 out of 83 fusion transcripts are private. In 322 endometrial carcinoma patients, 47 out of 49 fusion transcripts are non-recurrent. Sensitivity and specificity were also used to calculate the positive and negative likelihood ratios and diagnostic odds ratios. The DOR was equal to the LR+ divided by the LR2. The positive likelihood ratio, the ratio of true-positives to false-positives amongst all those coded for HF, was equal to the sensitivity divided by specificity. The negative likelihood ratio, the ratio of false-negatives to true-negatives amongst all those not coded for HF, was equal to sensitivity divided by the specificity. Thus, higher LR+ Butylhydroxyanisole values mean the presence of an HF code is more indicative of true HF and lower LR- values mean the absence of an HF code is more indicative of non-disease.