Bioinformatic Identification of a Breast-Specific Transcript Profile.

PURPOSE To identify a breast-specific transcript profile for the first time, and present an updated bioinformatics strategy for searching tissue-specific transcripts and predicting their significance in cancer. EXPERIMENTAL DESIGN The RNA-Seq data of 49,311 transcripts in 88 human tissues from the GTEx, the Illumina Body Map, and the RIKEN FANTOM5 project were integrated to screen breast-specific transcripts, and GEPIA, TGCA, and Kaplan-Meier Plotter were used to exam their expression in cancer tissues and values for prognosis prediction. RESULTS Only 96 transcripts in human genome were breast-specific for women. Among them, ANKRD30A and LINC00993 were further analyzed. The two transcripts were also breast-specific in 33 types of common female cancer and often dysregulated in breast cancer tissues. Their expression was higher in the luminal breast cancer while significantly down-regulated in triple-negative breast cancer. Moreover, the high expression levels of ANKRD30A and LINC0993 in breast cancer tissues indicated a better prognosis of patients with breast cancer. CONCLUSIONS AND CLINICAL RELEVANCE Breast-specific transcripts in human genome are rare and poorly understood currently. Our data indicated that these breast-specific biomarkers are promising candidates for screening early cancer, assessing treatment response, monitoring recurrence, identifying metastatic tumor origin, and serving as potential targets for immunotherapy. This article is protected by copyright. All rights reserved.