Improved detection of clinically relevant fusion transcripts in cancer by machine learning classification
Abstract Background Genomic rearrangements in cancer cells can create fusion genes that encode chimeric proteins or alter the expression of coding and non-coding RNAs. In some cancer types, fusions involving specific kinases are used as targets for therapy. Fusion genes can be detected by whole geno...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2023-12-01
|
| Series: | BMC Genomics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12864-023-09889-y |
| _version_ | 1797377213238083584 |
|---|---|
| author | Völundur Hafstað Jari Häkkinen Malin Larsson Johan Staaf Johan Vallon-Christersson Helena Persson |
| author_facet | Völundur Hafstað Jari Häkkinen Malin Larsson Johan Staaf Johan Vallon-Christersson Helena Persson |
| author_sort | Völundur Hafstað |
| collection | DOAJ |
| description | Abstract Background Genomic rearrangements in cancer cells can create fusion genes that encode chimeric proteins or alter the expression of coding and non-coding RNAs. In some cancer types, fusions involving specific kinases are used as targets for therapy. Fusion genes can be detected by whole genome sequencing (WGS) and targeted fusion panels, but RNA sequencing (RNA-Seq) has the advantageous capability of broadly detecting expressed fusion transcripts. Results We developed a pipeline for validation of fusion transcripts identified in RNA-Seq data using matched WGS data from The Cancer Genome Atlas (TCGA) and applied it to 910 tumors from 11 different cancer types. This resulted in 4237 validated gene fusions, 3049 of them with at least one identified genomic breakpoint. Utilizing validated fusions as true positive events, we trained a machine learning classifier to predict true and false positive fusion transcripts from RNA-Seq data. The final precision and recall metrics of the classifier were 0.74 and 0.71, respectively, in an independent dataset of 249 breast tumors. Application of this classifier to all samples with RNA-Seq data from these cancer types vastly extended the number of likely true positive fusion transcripts and identified many potentially targetable kinase fusions. Further analysis of the validated gene fusions suggested that many are created by intrachromosomal amplification events with microhomology-mediated non-homologous end-joining. Conclusions A classifier trained on validated fusion events increased the accuracy of fusion transcript identification in samples without WGS data. This allowed the analysis to be extended to all samples with RNA-Seq data, facilitating studies of tumor biology and increasing the number of detected kinase fusions. Machine learning could thus be used in identification of clinically relevant fusion events for targeted therapy. The large dataset of validated gene fusions generated here presents a useful resource for development and evaluation of fusion transcript detection algorithms. |
| first_indexed | 2024-03-08T19:49:32Z |
| format | Article |
| id | doaj.art-9ff2da0262514696b39df8415e9a8fba |
| institution | Directory Open Access Journal |
| issn | 1471-2164 |
| language | English |
| last_indexed | 2024-03-08T19:49:32Z |
| publishDate | 2023-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Genomics |
| spelling | doaj.art-9ff2da0262514696b39df8415e9a8fba2023-12-24T12:10:32ZengBMCBMC Genomics1471-21642023-12-0124111610.1186/s12864-023-09889-yImproved detection of clinically relevant fusion transcripts in cancer by machine learning classificationVölundur Hafstað0Jari Häkkinen1Malin Larsson2Johan Staaf3Johan Vallon-Christersson4Helena Persson5Faculty of Medicine, Department of Clinical Sciences Lund, Oncology, Lund University Cancer CentreFaculty of Medicine, Department of Clinical Sciences Lund, Oncology, Lund University Cancer CentreDepartment of Physics, Chemistry and Biology, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Linköping UniversityFaculty of Medicine, Department of Laboratory Medicine, Translational Cancer Research, Lund University Cancer CentreFaculty of Medicine, Department of Clinical Sciences Lund, Oncology, Lund University Cancer CentreFaculty of Medicine, Department of Clinical Sciences Lund, Oncology, Lund University Cancer CentreAbstract Background Genomic rearrangements in cancer cells can create fusion genes that encode chimeric proteins or alter the expression of coding and non-coding RNAs. In some cancer types, fusions involving specific kinases are used as targets for therapy. Fusion genes can be detected by whole genome sequencing (WGS) and targeted fusion panels, but RNA sequencing (RNA-Seq) has the advantageous capability of broadly detecting expressed fusion transcripts. Results We developed a pipeline for validation of fusion transcripts identified in RNA-Seq data using matched WGS data from The Cancer Genome Atlas (TCGA) and applied it to 910 tumors from 11 different cancer types. This resulted in 4237 validated gene fusions, 3049 of them with at least one identified genomic breakpoint. Utilizing validated fusions as true positive events, we trained a machine learning classifier to predict true and false positive fusion transcripts from RNA-Seq data. The final precision and recall metrics of the classifier were 0.74 and 0.71, respectively, in an independent dataset of 249 breast tumors. Application of this classifier to all samples with RNA-Seq data from these cancer types vastly extended the number of likely true positive fusion transcripts and identified many potentially targetable kinase fusions. Further analysis of the validated gene fusions suggested that many are created by intrachromosomal amplification events with microhomology-mediated non-homologous end-joining. Conclusions A classifier trained on validated fusion events increased the accuracy of fusion transcript identification in samples without WGS data. This allowed the analysis to be extended to all samples with RNA-Seq data, facilitating studies of tumor biology and increasing the number of detected kinase fusions. Machine learning could thus be used in identification of clinically relevant fusion events for targeted therapy. The large dataset of validated gene fusions generated here presents a useful resource for development and evaluation of fusion transcript detection algorithms.https://doi.org/10.1186/s12864-023-09889-yFusion transcriptGene fusionCancer genomicsTumor biologyPrecision medicineMachine learning |
| spellingShingle | Völundur Hafstað Jari Häkkinen Malin Larsson Johan Staaf Johan Vallon-Christersson Helena Persson Improved detection of clinically relevant fusion transcripts in cancer by machine learning classification BMC Genomics Fusion transcript Gene fusion Cancer genomics Tumor biology Precision medicine Machine learning |
| title | Improved detection of clinically relevant fusion transcripts in cancer by machine learning classification |
| title_full | Improved detection of clinically relevant fusion transcripts in cancer by machine learning classification |
| title_fullStr | Improved detection of clinically relevant fusion transcripts in cancer by machine learning classification |
| title_full_unstemmed | Improved detection of clinically relevant fusion transcripts in cancer by machine learning classification |
| title_short | Improved detection of clinically relevant fusion transcripts in cancer by machine learning classification |
| title_sort | improved detection of clinically relevant fusion transcripts in cancer by machine learning classification |
| topic | Fusion transcript Gene fusion Cancer genomics Tumor biology Precision medicine Machine learning |
| url | https://doi.org/10.1186/s12864-023-09889-y |
| work_keys_str_mv | AT volundurhafstað improveddetectionofclinicallyrelevantfusiontranscriptsincancerbymachinelearningclassification AT jarihakkinen improveddetectionofclinicallyrelevantfusiontranscriptsincancerbymachinelearningclassification AT malinlarsson improveddetectionofclinicallyrelevantfusiontranscriptsincancerbymachinelearningclassification AT johanstaaf improveddetectionofclinicallyrelevantfusiontranscriptsincancerbymachinelearningclassification AT johanvallonchristersson improveddetectionofclinicallyrelevantfusiontranscriptsincancerbymachinelearningclassification AT helenapersson improveddetectionofclinicallyrelevantfusiontranscriptsincancerbymachinelearningclassification |