This article is part of the supplement: Proceedings of the Eighth Annual MCBIOS Conference. Computational Biology and Bioinformatics for a New Decade
Selecting a single model or combining multiple models for microarray-based classifier development? – A comparative analysis based on large and diverse datasets generated from the MAQC-II project
1 Center for Bioinformatics, Division of Systems Biology, National Center for Toxicological Research, U.S. Food & Drug Administration, 3900 NCTR Rd, Jefferson, Arkansas, USA
2 ICF International at FDA's National Center for Toxicological Research, 3900 NCTR Rd, Jefferson, AR 72079, USA
BMC Bioinformatics 2011, 12(Suppl 10):S3 doi:10.1186/1471-2105-12-S10-S3Published: 18 October 2011
Additional file 1:
Internal cross-validation vs. external validation of the 8320 NCTR developed models. The Pearson correlation of MCCs from Internal cross-validation vs. external validation is 0.927.
Format: BMP Size: 23.7MB Download file
Additional file 2:
Internal cross-validation vs. external validation of the NCTR nominated models.
Format: BMP Size: 27.1MB Download file
Additional file 3:
The average MCCs vs. the percentages of the top models for ensemble calculation. The average MCC was calculated from 13 endpoints in the external validation set; the top models were selected based on the MCCs from internal cross-validation in the training set.
Format: BMP Size: 22.6MB Download file