Power in pairs: assessing the statistical value of paired samples in tests for differential expression
Abstract Background When genomics researchers design a high-throughput study to test for differential expression, some biological systems and research questions provide opportunities to use paired samples from subjects, and researchers can plan for a certain proportion of subjects to have paired sam...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2018-12-01
|
| Series: | BMC Genomics |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12864-018-5236-2 |
| _version_ | 1818684247149379584 |
|---|---|
| author | John R. Stevens Jennifer S. Herrick Roger K. Wolff Martha L. Slattery |
| author_facet | John R. Stevens Jennifer S. Herrick Roger K. Wolff Martha L. Slattery |
| author_sort | John R. Stevens |
| collection | DOAJ |
| description | Abstract Background When genomics researchers design a high-throughput study to test for differential expression, some biological systems and research questions provide opportunities to use paired samples from subjects, and researchers can plan for a certain proportion of subjects to have paired samples. We consider the effect of this paired samples proportion on the statistical power of the study, using characteristics of both count (RNA-Seq) and continuous (microarray) expression data from a colorectal cancer study. Results We demonstrate that a higher proportion of subjects with paired samples yields higher statistical power, for various total numbers of samples, and for various strengths of subject-level confounding factors. In the design scenarios considered, the statistical power in a fully-paired design is substantially (and in many cases several times) greater than in an unpaired design. Conclusions For the many biological systems and research questions where paired samples are feasible and relevant, substantial statistical power gains can be achieved at the study design stage when genomics researchers plan on using paired samples from the largest possible proportion of subjects. Any cost savings in a study design with unpaired samples are likely accompanied by underpowered and possibly biased results. |
| first_indexed | 2024-12-17T10:47:36Z |
| format | Article |
| id | doaj.art-f3e41dae97774203989294f3cf127fcb |
| institution | Directory Open Access Journal |
| issn | 1471-2164 |
| language | English |
| last_indexed | 2024-12-17T10:47:36Z |
| publishDate | 2018-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Genomics |
| spelling | doaj.art-f3e41dae97774203989294f3cf127fcb2022-12-21T21:52:05ZengBMCBMC Genomics1471-21642018-12-0119111310.1186/s12864-018-5236-2Power in pairs: assessing the statistical value of paired samples in tests for differential expressionJohn R. Stevens0Jennifer S. Herrick1Roger K. Wolff2Martha L. Slattery3Department of Mathematics and Statistics, Utah State UniversityDivision of Epidemiology, Department of Internal Medicine, University of UtahDivision of Epidemiology, Department of Internal Medicine, University of UtahDivision of Epidemiology, Department of Internal Medicine, University of UtahAbstract Background When genomics researchers design a high-throughput study to test for differential expression, some biological systems and research questions provide opportunities to use paired samples from subjects, and researchers can plan for a certain proportion of subjects to have paired samples. We consider the effect of this paired samples proportion on the statistical power of the study, using characteristics of both count (RNA-Seq) and continuous (microarray) expression data from a colorectal cancer study. Results We demonstrate that a higher proportion of subjects with paired samples yields higher statistical power, for various total numbers of samples, and for various strengths of subject-level confounding factors. In the design scenarios considered, the statistical power in a fully-paired design is substantially (and in many cases several times) greater than in an unpaired design. Conclusions For the many biological systems and research questions where paired samples are feasible and relevant, substantial statistical power gains can be achieved at the study design stage when genomics researchers plan on using paired samples from the largest possible proportion of subjects. Any cost savings in a study design with unpaired samples are likely accompanied by underpowered and possibly biased results.http://link.springer.com/article/10.1186/s12864-018-5236-2Study designStatistical powerRNA-SeqMicroarraymicroRNA |
| spellingShingle | John R. Stevens Jennifer S. Herrick Roger K. Wolff Martha L. Slattery Power in pairs: assessing the statistical value of paired samples in tests for differential expression BMC Genomics Study design Statistical power RNA-Seq Microarray microRNA |
| title | Power in pairs: assessing the statistical value of paired samples in tests for differential expression |
| title_full | Power in pairs: assessing the statistical value of paired samples in tests for differential expression |
| title_fullStr | Power in pairs: assessing the statistical value of paired samples in tests for differential expression |
| title_full_unstemmed | Power in pairs: assessing the statistical value of paired samples in tests for differential expression |
| title_short | Power in pairs: assessing the statistical value of paired samples in tests for differential expression |
| title_sort | power in pairs assessing the statistical value of paired samples in tests for differential expression |
| topic | Study design Statistical power RNA-Seq Microarray microRNA |
| url | http://link.springer.com/article/10.1186/s12864-018-5236-2 |
| work_keys_str_mv | AT johnrstevens powerinpairsassessingthestatisticalvalueofpairedsamplesintestsfordifferentialexpression AT jennifersherrick powerinpairsassessingthestatisticalvalueofpairedsamplesintestsfordifferentialexpression AT rogerkwolff powerinpairsassessingthestatisticalvalueofpairedsamplesintestsfordifferentialexpression AT marthalslattery powerinpairsassessingthestatisticalvalueofpairedsamplesintestsfordifferentialexpression |