Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selection
<p>Abstract</p> <p>Background</p> <p>Algorithms designed to detect complex genetic disease associations are initially evaluated using simulated datasets. Typical evaluations vary constraints that influence the correct detection of underlying models (i.e. number of loci,...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2012-09-01
|
| Series: | BioData Mining |
| Subjects: | |
| Online Access: | http://www.biodatamining.org/content/5/1/15 |
| _version_ | 1818835538634866688 |
|---|---|
| author | Urbanowicz Ryan J Kiralis Jeff Fisher Jonathan M Moore Jason H |
| author_facet | Urbanowicz Ryan J Kiralis Jeff Fisher Jonathan M Moore Jason H |
| author_sort | Urbanowicz Ryan J |
| collection | DOAJ |
| description | <p>Abstract</p> <p>Background</p> <p>Algorithms designed to detect complex genetic disease associations are initially evaluated using simulated datasets. Typical evaluations vary constraints that influence the correct detection of underlying models (i.e. number of loci, heritability, and minor allele frequency). Such studies neglect to account for model architecture (i.e. the unique specification and arrangement of penetrance values comprising the genetic model), which alone can influence the detectability of a model. In order to design a simulation study which efficiently takes architecture into account, a reliable metric is needed for model selection.</p> <p>Results</p> <p>We evaluate three metrics as predictors of relative model detection difficulty derived from previous works: (1) Penetrance table variance (PTV), (2) customized odds ratio (COR), and (3) our own Ease of Detection Measure (EDM), calculated from the penetrance values and respective genotype frequencies of each simulated genetic model. We evaluate the reliability of these metrics across three very different data search algorithms, each with the capacity to detect epistatic interactions. We find that a model’s EDM and COR are each stronger predictors of model detection success than heritability.</p> <p>Conclusions</p> <p>This study formally identifies and evaluates metrics which quantify model detection difficulty. We utilize these metrics to intelligently select models from a population of potential architectures. This allows for an improved simulation study design which accounts for differences in detection difficulty attributed to model architecture. We implement the calculation and utilization of EDM and COR into GAMETES, an algorithm which rapidly and precisely generates pure, strict, <it>n</it>-locus epistatic models.</p> |
| first_indexed | 2024-12-19T02:52:18Z |
| format | Article |
| id | doaj.art-e2bfeff8932340e28ecd2bab92c9d349 |
| institution | Directory Open Access Journal |
| issn | 1756-0381 |
| language | English |
| last_indexed | 2024-12-19T02:52:18Z |
| publishDate | 2012-09-01 |
| publisher | BMC |
| record_format | Article |
| series | BioData Mining |
| spelling | doaj.art-e2bfeff8932340e28ecd2bab92c9d3492022-12-21T20:38:35ZengBMCBioData Mining1756-03812012-09-01511510.1186/1756-0381-5-15Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selectionUrbanowicz Ryan JKiralis JeffFisher Jonathan MMoore Jason H<p>Abstract</p> <p>Background</p> <p>Algorithms designed to detect complex genetic disease associations are initially evaluated using simulated datasets. Typical evaluations vary constraints that influence the correct detection of underlying models (i.e. number of loci, heritability, and minor allele frequency). Such studies neglect to account for model architecture (i.e. the unique specification and arrangement of penetrance values comprising the genetic model), which alone can influence the detectability of a model. In order to design a simulation study which efficiently takes architecture into account, a reliable metric is needed for model selection.</p> <p>Results</p> <p>We evaluate three metrics as predictors of relative model detection difficulty derived from previous works: (1) Penetrance table variance (PTV), (2) customized odds ratio (COR), and (3) our own Ease of Detection Measure (EDM), calculated from the penetrance values and respective genotype frequencies of each simulated genetic model. We evaluate the reliability of these metrics across three very different data search algorithms, each with the capacity to detect epistatic interactions. We find that a model’s EDM and COR are each stronger predictors of model detection success than heritability.</p> <p>Conclusions</p> <p>This study formally identifies and evaluates metrics which quantify model detection difficulty. We utilize these metrics to intelligently select models from a population of potential architectures. This allows for an improved simulation study design which accounts for differences in detection difficulty attributed to model architecture. We implement the calculation and utilization of EDM and COR into GAMETES, an algorithm which rapidly and precisely generates pure, strict, <it>n</it>-locus epistatic models.</p>http://www.biodatamining.org/content/5/1/15EDMCORGAMETESSNPModel detectionEpistasisSimulationModelGenetics |
| spellingShingle | Urbanowicz Ryan J Kiralis Jeff Fisher Jonathan M Moore Jason H Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selection BioData Mining EDM COR GAMETES SNP Model detection Epistasis Simulation Model Genetics |
| title | Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selection |
| title_full | Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selection |
| title_fullStr | Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selection |
| title_full_unstemmed | Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selection |
| title_short | Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selection |
| title_sort | predicting the difficulty of pure strict epistatic models metrics for simulated model selection |
| topic | EDM COR GAMETES SNP Model detection Epistasis Simulation Model Genetics |
| url | http://www.biodatamining.org/content/5/1/15 |
| work_keys_str_mv | AT urbanowiczryanj predictingthedifficultyofpurestrictepistaticmodelsmetricsforsimulatedmodelselection AT kiralisjeff predictingthedifficultyofpurestrictepistaticmodelsmetricsforsimulatedmodelselection AT fisherjonathanm predictingthedifficultyofpurestrictepistaticmodelsmetricsforsimulatedmodelselection AT moorejasonh predictingthedifficultyofpurestrictepistaticmodelsmetricsforsimulatedmodelselection |