Limits...
Integrated RNA and DNA sequencing improves mutation detection in low purity tumors.

Wilkerson MD, Cabanski CR, Sun W, Hoadley KA, Walter V, Mose LE, Troester MA, Hammerman PS, Parker JS, Perou CM, Hayes DN - Nucleic Acids Res. (2014)

Bottom Line: We hypothesized that integrating DNA-WES and RNA-seq would enable superior mutation detection versus DNA-WES alone.Validation by patient-matched whole genome sequencing demonstrated superior performance of the integrated model over DNA-WES only models, including a published method and published mutation profiles.RNA provided greater mutation signal than DNA in expressed mutations.

View Article: PubMed Central - PubMed

Affiliation: Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA Department of Genetics, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA mwilkers@med.unc.edu.

Show MeSH

Related in: MedlinePlus

Validation of mutation detection by whole genome sequencing. The number of true positives and false positives of mutation detection models are plotted as step functions. At fixed false positive totals (250, 500 or 1000), each pair of models was compared for differences in number of true positives (*). The published mutation set (4,6) did not include mutation rankings and was not amenable to rank-based statistical analysis.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4117748&req=5

Figure 2: Validation of mutation detection by whole genome sequencing. The number of true positives and false positives of mutation detection models are plotted as step functions. At fixed false positive totals (250, 500 or 1000), each pair of models was compared for differences in number of true positives (*). The published mutation set (4,6) did not include mutation rankings and was not amenable to rank-based statistical analysis.

Mentions: To validate the superior performance of integrated DNA-WES and RNA-seq mutation detection (UNCeqRMETA) over DNA-WES only detection (UNCeqRDNA), tumor and germline whole genome DNA sequencing (DNA-WGS) was used as an independent measure of truth for evaluating DNA-WES and RNA-seq mutation detections. Following a published validation procedure (4), mutation detections were interrogated in patient-matched DNA-WGS to determine if a mutation detection was a true positive, that is present in the tumor specimen and absent from the germline specimen, or false positive, that is absent from the tumor specimen or present in the germline specimen. For each mutation model, true positives and false positives were summed at each discrimination threshold (e.g. P-value) to generate a performance curve by which true positive rates could be compared at the same false positive rates (see methods for further description). These curves demonstrated that UNCeqRMETA achieved overall superior performance than UNCeqRDNA (difference in area under the curve, P < 0.01) and at fixed false positive thresholds (250, 500 and 1000), thus, validating the result from simulated tumor genomes (Figure 2). Therefore, in real tumor sequencing, integrated DNA and RNA mutation detection by UNCeqRMETA outperformed DNA-only mutation detection.


Integrated RNA and DNA sequencing improves mutation detection in low purity tumors.

Wilkerson MD, Cabanski CR, Sun W, Hoadley KA, Walter V, Mose LE, Troester MA, Hammerman PS, Parker JS, Perou CM, Hayes DN - Nucleic Acids Res. (2014)

Validation of mutation detection by whole genome sequencing. The number of true positives and false positives of mutation detection models are plotted as step functions. At fixed false positive totals (250, 500 or 1000), each pair of models was compared for differences in number of true positives (*). The published mutation set (4,6) did not include mutation rankings and was not amenable to rank-based statistical analysis.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4117748&req=5

Figure 2: Validation of mutation detection by whole genome sequencing. The number of true positives and false positives of mutation detection models are plotted as step functions. At fixed false positive totals (250, 500 or 1000), each pair of models was compared for differences in number of true positives (*). The published mutation set (4,6) did not include mutation rankings and was not amenable to rank-based statistical analysis.
Mentions: To validate the superior performance of integrated DNA-WES and RNA-seq mutation detection (UNCeqRMETA) over DNA-WES only detection (UNCeqRDNA), tumor and germline whole genome DNA sequencing (DNA-WGS) was used as an independent measure of truth for evaluating DNA-WES and RNA-seq mutation detections. Following a published validation procedure (4), mutation detections were interrogated in patient-matched DNA-WGS to determine if a mutation detection was a true positive, that is present in the tumor specimen and absent from the germline specimen, or false positive, that is absent from the tumor specimen or present in the germline specimen. For each mutation model, true positives and false positives were summed at each discrimination threshold (e.g. P-value) to generate a performance curve by which true positive rates could be compared at the same false positive rates (see methods for further description). These curves demonstrated that UNCeqRMETA achieved overall superior performance than UNCeqRDNA (difference in area under the curve, P < 0.01) and at fixed false positive thresholds (250, 500 and 1000), thus, validating the result from simulated tumor genomes (Figure 2). Therefore, in real tumor sequencing, integrated DNA and RNA mutation detection by UNCeqRMETA outperformed DNA-only mutation detection.

Bottom Line: We hypothesized that integrating DNA-WES and RNA-seq would enable superior mutation detection versus DNA-WES alone.Validation by patient-matched whole genome sequencing demonstrated superior performance of the integrated model over DNA-WES only models, including a published method and published mutation profiles.RNA provided greater mutation signal than DNA in expressed mutations.

View Article: PubMed Central - PubMed

Affiliation: Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA Department of Genetics, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA mwilkers@med.unc.edu.

Show MeSH
Related in: MedlinePlus