Dataset Information

InFusion: advancing discovery of fusion genes and chimeric transcripts from RNA-seq data

ABSTRACT: Gene fusions and chimeric transcripts occur frequently in cancers and in some cases drive the development of the disease. An accurate detection of these events is crucial for cancer research and in a long-term perspective could be applied for personalized therapy. RNA-seq technology has been established as an efficient approach to investigate transcriptomes and search for gene fusions and chimeric transcripts on a genome-wide scale. A number of computational methods for the detection of gene fusions from RNA-seq data have been developed. However, recent studies demonstrate differences between commonly used approaches in terms of specificity and sensitivity. Moreover their ability to detect gene fusions on the isoform level has not been studied carefully so far. Here we propose a novel computational approach called InFusion for fusion gene detection from deep RNA sequencing data. Validation of InFusion on simulated and on several public RNA-seq datasets demonstrated better detection accuracy compared to other tools. We also performed deep RNA sequencing of two well-established prostate cancer cell lines. Using these data we showed that InFusion is capable of discovering alternatively spliced gene fusion isoforms as well as chimeric transcripts that include non-exonic regions. In addition our method can detect anti-sense transcription in the fusions by incorporating strand specificity of the sequencing library.

ORGANISM(S): Homo sapiens

PROVIDER: GSE56512 | GEO | 2014/12/01

SECONDARY ACCESSION(S): PRJNA243568

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Purpose: Assessment of the performance characteristics of an RNA-Seq assay designed to detect gene fusions in 573 genes to aid in the management of cancer patients. Methods: Polyadenylated RNA was converted to cDNA which was then used to prepare NGS libraries that were sequenced on a HiSeq 2500 instrument and analyzed with an in-house developed bioinformatic pipeline. Results: The assay identified 38 of 41 (93%) gene fusions previously detected by a different laboratory using FISH, RT-PCR, or RNA-Seq for a sensitivity of 93%. No false positive gene fusions were identified in 15 normal tissue specimens and 10 tumor specimens that were negative for fusions by RNA-Seq in a different laboratory (100% specificity). The assay also identified 22 fusions in 17 tumor specimens that had not been detected by other methods. Nineteen of the 22 fusions had not previously been described. Good intra- and inter-assay reproducibility was observed with complete concordance for the presence or absence of gene fusions in replicates. The analytical sensitivity of the assay was tested by diluting RNA isolated from gene fusion positive cases with fusion negative RNA. Gene fusions were generally detectable down to 12.5% dilutions for most fusions and as little as 3% for some fusions. The assay identified 38 of 41 (93%) gene fusions previously detected by a different laboratory using FISH, RT-PCR, or RNA-Seq for a sensitivity of 93%. No false positive gene fusions were identified in 15 normal tissue specimens and 10 tumor specimens that were negative for fusions by RNA-Seq in a different laboratory (100% specificity). The assay also identified 22 fusions in 17 tumor specimens that had not been detected by other methods. Nineteen of the 22 fusions had not previously been described. Good intra- and inter-assay reproducibility was observed with complete concordance for the presence or absence of gene fusions in replicates. The analytical sensitivity of the assay was tested by diluting RNA isolated from gene fusion positive cases with fusion negative RNA. Gene fusions were generally detectable down to 12.5% dilutions for most fusions and as little as 3% for some fusions. This assay should be useful for identifying cancer patients that may benefit from both FDA-approved and investigational targeted therapies.

Dataset Information

InFusion: advancing discovery of fusion genes and chimeric transcripts from RNA-seq data

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure