Limits...
An Efficient Method for Identifying Gene Fusions by Targeted RNA Sequencing from Fresh Frozen and FFPE Samples.

Scolnick JA, Dimon M, Wang IC, Huelga SC, Amorese DA - PLoS ONE (2015)

Bottom Line: Traditionally, detecting fusion genes has been a difficult task based on fluorescent in situ hybridization to detect chromosomal abnormalities.More recently, RNA sequencing has enabled an increased pace of fusion gene identification.Here we describe a method, Single Primer Enrichment Technology (SPET), for targeted RNA sequencing that is customizable to any target genes, is simple to use, and efficiently detects gene fusions.

View Article: PubMed Central - PubMed

Affiliation: NuGEN Technologies, Inc., San Carlos, California, United States of America.

ABSTRACT
Fusion genes are known to be key drivers of tumor growth in several types of cancer. Traditionally, detecting fusion genes has been a difficult task based on fluorescent in situ hybridization to detect chromosomal abnormalities. More recently, RNA sequencing has enabled an increased pace of fusion gene identification. However, RNA-Seq is inefficient for the identification of fusion genes due to the high number of sequencing reads needed to detect the small number of fusion transcripts present in cells of interest. Here we describe a method, Single Primer Enrichment Technology (SPET), for targeted RNA sequencing that is customizable to any target genes, is simple to use, and efficiently detects gene fusions. Using SPET to target 5701 exons of 401 known cancer fusion genes for sequencing, we were able to identify known and previously unreported gene fusions from both fresh-frozen and formalin-fixed paraffin-embedded (FFPE) tissue RNA in both normal tissue and cancer cells.

No MeSH data available.


Related in: MedlinePlus

Ovation Fusion Panel Target Enrichment System identifies known and novel gene fusion events in Universal Human Reference RNA.The number of sequencing reads determined to be derived from gene fusions in two different targeted sequencing libraries (blue and green) compared to the events identified in a standard, untargeted RNA-Seq library. The untargeted library (red) consists of 125 million total sequencing reads while the targeted libraries consist of 1.6 and 8.7 million sequencing reads.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4488430&req=5

pone.0128916.g002: Ovation Fusion Panel Target Enrichment System identifies known and novel gene fusion events in Universal Human Reference RNA.The number of sequencing reads determined to be derived from gene fusions in two different targeted sequencing libraries (blue and green) compared to the events identified in a standard, untargeted RNA-Seq library. The untargeted library (red) consists of 125 million total sequencing reads while the targeted libraries consist of 1.6 and 8.7 million sequencing reads.

Mentions: We next wanted to compare our results to those of standard RNA-Seq experiments to test how many sequencing reads were necessary to identify gene fusion events when using the Ovation Fusion Panel Target Enrichment System relative to a non-targeted RNA-Seq approach. Universal Human Reference RNA (UHR) contains RNA from ten different cancer cell lines. Many standard RNA-Seq experiments have been performed using UHR RNA and these experiments have identified a number of gene fusion events [8]. Our 401 gene panel contains probes that target two of the fusions previously shown to exist in UHR RNA, BCR-ABL and NUP214-XKR3. 100 ng of UHR total RNA was used to prepare sequencing libraries with Ovation Fusion Panel Target Enrichment System. Libraries were sequenced on either the Illumina MiSeq or HiSeq and analyzed with SOAPFuse [6]. Our initial tests utilizing 1.6M sequencing reads resulted in the identification of both BCR-ABL and NUP214-XKR3 fusion transcripts. In addition, we also identified an FGFR1-WHSC1L1 gene fusion, which has been reported as a recurring fusion in breast cancer [9], but has not been reported in UHR RNA. Deeper sequencing of another library to 8.7M reads on the Illumina HiSeq led to the further identification of another previously unreported UHR gene fusion, MLLT10 fused to PICALM. Having identified a set of gene fusions in UHR RNA, we compared the number of fusion reads in these genes detected by SOAPFuse in our datasets to those of a publically available RNA-Seq dataset that consists of 125M paired end sequencing reads [10]. As seen in Fig 2, when comparing 1.6M Ovation Fusion Panel Target Enrichment System sequencing reads (blue bars) to 125M standard RNA-Seq sequencing reads (red bars), fusion read counts are comparable for the BCR-ABL fusion. The 1.6M targeted sequencing reads has fewer reads for NUP214-XKR3, which is expected because our targeting panel only contains probes for NUP214 and not XKR3. The standard RNA-Seq does not contain any detectable reads for FGFR1-WHSC1L1 nor any reads for MLLT10-PICALM, while the targeted sequencing approach described here identified both of these fusion events. Both of these previously undescribed fusions were verified by RT-PCR and Sanger sequencing. In summary, by using a targeted sequencing approach, we were able to identify known gene fusion events in UHR RNA while using only 1.3% of the sequencing reads necessary to identify approximately the same number of fusion reads in a standard RNA sequencing experiment. In addition we were able to identify previously undocumented gene fusion events using the targeted approach that were not identified in standard RNA sequencing, even though the targeted sequencing had 14–78X fewer sequencing reads. The inability of standard RNA sequencing to identify fusion transcripts in a cell line that has been studied in detail suggests that even samples for which deep RNA sequencing data already exists should be tested again specifically for the presence of fusion transcripts.


An Efficient Method for Identifying Gene Fusions by Targeted RNA Sequencing from Fresh Frozen and FFPE Samples.

Scolnick JA, Dimon M, Wang IC, Huelga SC, Amorese DA - PLoS ONE (2015)

Ovation Fusion Panel Target Enrichment System identifies known and novel gene fusion events in Universal Human Reference RNA.The number of sequencing reads determined to be derived from gene fusions in two different targeted sequencing libraries (blue and green) compared to the events identified in a standard, untargeted RNA-Seq library. The untargeted library (red) consists of 125 million total sequencing reads while the targeted libraries consist of 1.6 and 8.7 million sequencing reads.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4488430&req=5

pone.0128916.g002: Ovation Fusion Panel Target Enrichment System identifies known and novel gene fusion events in Universal Human Reference RNA.The number of sequencing reads determined to be derived from gene fusions in two different targeted sequencing libraries (blue and green) compared to the events identified in a standard, untargeted RNA-Seq library. The untargeted library (red) consists of 125 million total sequencing reads while the targeted libraries consist of 1.6 and 8.7 million sequencing reads.
Mentions: We next wanted to compare our results to those of standard RNA-Seq experiments to test how many sequencing reads were necessary to identify gene fusion events when using the Ovation Fusion Panel Target Enrichment System relative to a non-targeted RNA-Seq approach. Universal Human Reference RNA (UHR) contains RNA from ten different cancer cell lines. Many standard RNA-Seq experiments have been performed using UHR RNA and these experiments have identified a number of gene fusion events [8]. Our 401 gene panel contains probes that target two of the fusions previously shown to exist in UHR RNA, BCR-ABL and NUP214-XKR3. 100 ng of UHR total RNA was used to prepare sequencing libraries with Ovation Fusion Panel Target Enrichment System. Libraries were sequenced on either the Illumina MiSeq or HiSeq and analyzed with SOAPFuse [6]. Our initial tests utilizing 1.6M sequencing reads resulted in the identification of both BCR-ABL and NUP214-XKR3 fusion transcripts. In addition, we also identified an FGFR1-WHSC1L1 gene fusion, which has been reported as a recurring fusion in breast cancer [9], but has not been reported in UHR RNA. Deeper sequencing of another library to 8.7M reads on the Illumina HiSeq led to the further identification of another previously unreported UHR gene fusion, MLLT10 fused to PICALM. Having identified a set of gene fusions in UHR RNA, we compared the number of fusion reads in these genes detected by SOAPFuse in our datasets to those of a publically available RNA-Seq dataset that consists of 125M paired end sequencing reads [10]. As seen in Fig 2, when comparing 1.6M Ovation Fusion Panel Target Enrichment System sequencing reads (blue bars) to 125M standard RNA-Seq sequencing reads (red bars), fusion read counts are comparable for the BCR-ABL fusion. The 1.6M targeted sequencing reads has fewer reads for NUP214-XKR3, which is expected because our targeting panel only contains probes for NUP214 and not XKR3. The standard RNA-Seq does not contain any detectable reads for FGFR1-WHSC1L1 nor any reads for MLLT10-PICALM, while the targeted sequencing approach described here identified both of these fusion events. Both of these previously undescribed fusions were verified by RT-PCR and Sanger sequencing. In summary, by using a targeted sequencing approach, we were able to identify known gene fusion events in UHR RNA while using only 1.3% of the sequencing reads necessary to identify approximately the same number of fusion reads in a standard RNA sequencing experiment. In addition we were able to identify previously undocumented gene fusion events using the targeted approach that were not identified in standard RNA sequencing, even though the targeted sequencing had 14–78X fewer sequencing reads. The inability of standard RNA sequencing to identify fusion transcripts in a cell line that has been studied in detail suggests that even samples for which deep RNA sequencing data already exists should be tested again specifically for the presence of fusion transcripts.

Bottom Line: Traditionally, detecting fusion genes has been a difficult task based on fluorescent in situ hybridization to detect chromosomal abnormalities.More recently, RNA sequencing has enabled an increased pace of fusion gene identification.Here we describe a method, Single Primer Enrichment Technology (SPET), for targeted RNA sequencing that is customizable to any target genes, is simple to use, and efficiently detects gene fusions.

View Article: PubMed Central - PubMed

Affiliation: NuGEN Technologies, Inc., San Carlos, California, United States of America.

ABSTRACT
Fusion genes are known to be key drivers of tumor growth in several types of cancer. Traditionally, detecting fusion genes has been a difficult task based on fluorescent in situ hybridization to detect chromosomal abnormalities. More recently, RNA sequencing has enabled an increased pace of fusion gene identification. However, RNA-Seq is inefficient for the identification of fusion genes due to the high number of sequencing reads needed to detect the small number of fusion transcripts present in cells of interest. Here we describe a method, Single Primer Enrichment Technology (SPET), for targeted RNA sequencing that is customizable to any target genes, is simple to use, and efficiently detects gene fusions. Using SPET to target 5701 exons of 401 known cancer fusion genes for sequencing, we were able to identify known and previously unreported gene fusions from both fresh-frozen and formalin-fixed paraffin-embedded (FFPE) tissue RNA in both normal tissue and cancer cells.

No MeSH data available.


Related in: MedlinePlus