Entropy-based gene ranking without selection bias for the predictive classification of microarray data
<p>Abstract</p> <p>Background</p> <p>We describe the E-RFE method for gene ranking, which is useful for the identification of markers in the predictive classification of array data. The method supports a practical modeling scheme designed to avoid the construction of cl...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2003-11-01
|
| Series: | BMC Bioinformatics |
| Online Access: | http://www.biomedcentral.com/1471-2105/4/54 |
| _version_ | 1818471645266837504 |
|---|---|
| author | Serafini Maria Furlanello Cesare Merler Stefano Jurman Giuseppe |
| author_facet | Serafini Maria Furlanello Cesare Merler Stefano Jurman Giuseppe |
| author_sort | Serafini Maria |
| collection | DOAJ |
| description | <p>Abstract</p> <p>Background</p> <p>We describe the E-RFE method for gene ranking, which is useful for the identification of markers in the predictive classification of array data. The method supports a practical modeling scheme designed to avoid the construction of classification rules based on the selection of too small gene subsets (an effect known as the selection bias, in which the estimated predictive errors are too optimistic due to testing on samples already considered in the feature selection process).</p> <p>Results</p> <p>With E-RFE, we speed up the recursive feature elimination (RFE) with SVM classifiers by eliminating chunks of uninteresting genes using an entropy measure of the SVM weights distribution. An optimal subset of genes is selected according to a two-strata model evaluation procedure: modeling is replicated by an external stratified-partition resampling scheme, and, within each run, an internal K-fold cross-validation is used for E-RFE ranking. Also, the optimal number of genes can be estimated according to the saturation of Zipf's law profiles.</p> <p>Conclusions</p> <p>Without a decrease of classification accuracy, E-RFE allows a speed-up factor of 100 with respect to standard RFE, while improving on alternative parametric RFE reduction strategies. Thus, a process for gene selection and error estimation is made practical, ensuring control of the selection bias, and providing additional diagnostic indicators of gene importance.</p> |
| first_indexed | 2024-04-14T03:54:12Z |
| format | Article |
| id | doaj.art-3598abb28f1a4366a007c8b087f25bdf |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-04-14T03:54:12Z |
| publishDate | 2003-11-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-3598abb28f1a4366a007c8b087f25bdf2022-12-22T02:13:52ZengBMCBMC Bioinformatics1471-21052003-11-01415410.1186/1471-2105-4-54Entropy-based gene ranking without selection bias for the predictive classification of microarray dataSerafini MariaFurlanello CesareMerler StefanoJurman Giuseppe<p>Abstract</p> <p>Background</p> <p>We describe the E-RFE method for gene ranking, which is useful for the identification of markers in the predictive classification of array data. The method supports a practical modeling scheme designed to avoid the construction of classification rules based on the selection of too small gene subsets (an effect known as the selection bias, in which the estimated predictive errors are too optimistic due to testing on samples already considered in the feature selection process).</p> <p>Results</p> <p>With E-RFE, we speed up the recursive feature elimination (RFE) with SVM classifiers by eliminating chunks of uninteresting genes using an entropy measure of the SVM weights distribution. An optimal subset of genes is selected according to a two-strata model evaluation procedure: modeling is replicated by an external stratified-partition resampling scheme, and, within each run, an internal K-fold cross-validation is used for E-RFE ranking. Also, the optimal number of genes can be estimated according to the saturation of Zipf's law profiles.</p> <p>Conclusions</p> <p>Without a decrease of classification accuracy, E-RFE allows a speed-up factor of 100 with respect to standard RFE, while improving on alternative parametric RFE reduction strategies. Thus, a process for gene selection and error estimation is made practical, ensuring control of the selection bias, and providing additional diagnostic indicators of gene importance.</p>http://www.biomedcentral.com/1471-2105/4/54 |
| spellingShingle | Serafini Maria Furlanello Cesare Merler Stefano Jurman Giuseppe Entropy-based gene ranking without selection bias for the predictive classification of microarray data BMC Bioinformatics |
| title | Entropy-based gene ranking without selection bias for the predictive classification of microarray data |
| title_full | Entropy-based gene ranking without selection bias for the predictive classification of microarray data |
| title_fullStr | Entropy-based gene ranking without selection bias for the predictive classification of microarray data |
| title_full_unstemmed | Entropy-based gene ranking without selection bias for the predictive classification of microarray data |
| title_short | Entropy-based gene ranking without selection bias for the predictive classification of microarray data |
| title_sort | entropy based gene ranking without selection bias for the predictive classification of microarray data |
| url | http://www.biomedcentral.com/1471-2105/4/54 |
| work_keys_str_mv | AT serafinimaria entropybasedgenerankingwithoutselectionbiasforthepredictiveclassificationofmicroarraydata AT furlanellocesare entropybasedgenerankingwithoutselectionbiasforthepredictiveclassificationofmicroarraydata AT merlerstefano entropybasedgenerankingwithoutselectionbiasforthepredictiveclassificationofmicroarraydata AT jurmangiuseppe entropybasedgenerankingwithoutselectionbiasforthepredictiveclassificationofmicroarraydata |