Abstract Fusion circular RNAs (f-circRNAs) are circular RNAs derived from rearranged genome translocations. Their formation is caused by back-splicing happening to the genome regions of aberrant chromosomal translocation. Their cancer-promoting roles have been implicated in recent studies. One of them adopted both in vivo and in vitro models to demonstrate the oncogenic roles of f-circRNAs including promoting cellular transformation, cell viability and resistance to treatment. Therefore, it is suggested that accurate identification of f-circRNA will contribute to cancer diagnosis and therapy. But until now there is only one tool called Acfs (accurate circRNA finder suite) ever mentioned detecting f-circRNAs whose primary job is to detect normal circRNAs and only 10 f-circRNAs have been tested for verification without considering realistic sequencing problems. Here, we proposed the first algorithm, FCRF (Fusion CircRNA Finder), to identify f-circRNAs from RNA-Seq data. The SAM file aligned by BWA was used as input, and paired fusion junction site locations would be output. After pre-processing and correcting the original data in consideration of the sequencing fault tolerance, we divided the candidates into balance set and unbalance set according to the relative aligned length. For balance set, we extracted reads from non-homologous chromosomes owing to the fusion events and further detected the paired clipping signals specific to fusion junctions. On the other hand, for unbalance set, we could not use the paired signal because BWA only reported one alignment for them. So we handled them with local similar sequence considering that each junction read has similar sequences with the other read spanning the same junction. Finally we searched all the candidate fusion junctions in accord with the threshold window size and cyclization characteristics, and then recorded two special junction site locations for each fusion-circRNA. We conducted a series of simulation experiments to evaluate our algorithm. The proposed algorithm FCRF achieved 95% on precision and 85% on sensitivity where Acfs output nothing from our improved dataset. Note: This abstract was not presented at the meeting. Citation Format: Han Yu, Xuanping Zhang, Yanfang Guan, Yongsheng Chen, Xiao Xiao, Jiayin Wang. FCRF: An efficient algorithm for detecting circular fusion transcript from RNA-Seq data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-212.