Limits...
Iterative Bayesian Model Averaging: a method for the application of survival analysis to high-dimensional microarray data.

Annest A, Bumgarner RE, Raftery AE, Yeung KY - BMC Bioinformatics (2009)

Bottom Line: Moreover, we achieved comparable results using only the 5 top selected genes with 100% posterior probabilities.Once again, we assigned the patients in the validation set to significantly distinct risk groups (p-value = 0.00139).The results from this study demonstrate that our procedure selects a small number of genes while eclipsing other methods in predictive performance, making it a highly accurate and cost-effective prognostic tool in the clinical setting.

View Article: PubMed Central - HTML - PubMed

Affiliation: Institute of Technology/Computing and Software Systems, University of Washington, Tacoma, WA 98402, USA. amanu@u.washington.edu

ABSTRACT

Background: Microarray technology is increasingly used to identify potential biomarkers for cancer prognostics and diagnostics. Previously, we have developed the iterative Bayesian Model Averaging (BMA) algorithm for use in classification. Here, we extend the iterative BMA algorithm for application to survival analysis on high-dimensional microarray data. The main goal in applying survival analysis to microarray data is to determine a highly predictive model of patients' time to event (such as death, relapse, or metastasis) using a small number of selected genes. Our multivariate procedure combines the effectiveness of multiple contending models by calculating the weighted average of their posterior probability distributions. Our results demonstrate that our iterative BMA algorithm for survival analysis achieves high prediction accuracy while consistently selecting a small and cost-effective number of predictor genes.

Results: We applied the iterative BMA algorithm to two cancer datasets: breast cancer and diffuse large B-cell lymphoma (DLBCL) data. On the breast cancer data, the algorithm selected a total of 15 predictor genes across 84 contending models from the training data. The maximum likelihood estimates of the selected genes and the posterior probabilities of the selected models from the training data were used to divide patients in the test (or validation) dataset into high- and low-risk categories. Using the genes and models determined from the training data, we assigned patients from the test data into highly distinct risk groups (as indicated by a p-value of 7.26e-05 from the log-rank test). Moreover, we achieved comparable results using only the 5 top selected genes with 100% posterior probabilities. On the DLBCL data, our iterative BMA procedure selected a total of 25 genes across 3 contending models from the training data. Once again, we assigned the patients in the validation set to significantly distinct risk groups (p-value = 0.00139).

Conclusion: The strength of the iterative BMA algorithm for survival analysis lies in its ability to account for model uncertainty. The results from this study demonstrate that our procedure selects a small number of genes while eclipsing other methods in predictive performance, making it a highly accurate and cost-effective prognostic tool in the clinical setting.

Show MeSH

Related in: MedlinePlus

Outline of the iterative BMA algorithm for survival analysis on microarray data.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2657791&req=5

Figure 1: Outline of the iterative BMA algorithm for survival analysis on microarray data.

Mentions: Following this step, the algorithm iterates through the user-specified p top-ranked genes, applying the traditional BMA algorithm for survival analysis [31] to each group of variables in the current BMA window (where the window size is denoted by maxNvar). This part of the procedure is similar to the classification method described previously; genes with high posterior probabilities are retained while genes with low posterior probabilities are eliminated. Following Yeung et al. [19], we have chosen to adopt the 1% default threshold for inclusion. The algorithm relies on the elimination of at least one gene per iteration from the current BMA window, so the method cannot proceed if all genes in the window have a posterior probability ≥ 1%. Yeung et al. proposed an "adaptive threshold" heuristic to account for this possibility, whereby the genes with the lowest posterior probabilities are removed to make room for subsequent variables. We have incorporated this heuristic into our algorithm because Yeung et al. [19] reported that its inclusion boosts predictive accuracy. See Figure 1 for an outline of the iterative BMA algorithm for survival analysis.


Iterative Bayesian Model Averaging: a method for the application of survival analysis to high-dimensional microarray data.

Annest A, Bumgarner RE, Raftery AE, Yeung KY - BMC Bioinformatics (2009)

Outline of the iterative BMA algorithm for survival analysis on microarray data.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2657791&req=5

Figure 1: Outline of the iterative BMA algorithm for survival analysis on microarray data.
Mentions: Following this step, the algorithm iterates through the user-specified p top-ranked genes, applying the traditional BMA algorithm for survival analysis [31] to each group of variables in the current BMA window (where the window size is denoted by maxNvar). This part of the procedure is similar to the classification method described previously; genes with high posterior probabilities are retained while genes with low posterior probabilities are eliminated. Following Yeung et al. [19], we have chosen to adopt the 1% default threshold for inclusion. The algorithm relies on the elimination of at least one gene per iteration from the current BMA window, so the method cannot proceed if all genes in the window have a posterior probability ≥ 1%. Yeung et al. proposed an "adaptive threshold" heuristic to account for this possibility, whereby the genes with the lowest posterior probabilities are removed to make room for subsequent variables. We have incorporated this heuristic into our algorithm because Yeung et al. [19] reported that its inclusion boosts predictive accuracy. See Figure 1 for an outline of the iterative BMA algorithm for survival analysis.

Bottom Line: Moreover, we achieved comparable results using only the 5 top selected genes with 100% posterior probabilities.Once again, we assigned the patients in the validation set to significantly distinct risk groups (p-value = 0.00139).The results from this study demonstrate that our procedure selects a small number of genes while eclipsing other methods in predictive performance, making it a highly accurate and cost-effective prognostic tool in the clinical setting.

View Article: PubMed Central - HTML - PubMed

Affiliation: Institute of Technology/Computing and Software Systems, University of Washington, Tacoma, WA 98402, USA. amanu@u.washington.edu

ABSTRACT

Background: Microarray technology is increasingly used to identify potential biomarkers for cancer prognostics and diagnostics. Previously, we have developed the iterative Bayesian Model Averaging (BMA) algorithm for use in classification. Here, we extend the iterative BMA algorithm for application to survival analysis on high-dimensional microarray data. The main goal in applying survival analysis to microarray data is to determine a highly predictive model of patients' time to event (such as death, relapse, or metastasis) using a small number of selected genes. Our multivariate procedure combines the effectiveness of multiple contending models by calculating the weighted average of their posterior probability distributions. Our results demonstrate that our iterative BMA algorithm for survival analysis achieves high prediction accuracy while consistently selecting a small and cost-effective number of predictor genes.

Results: We applied the iterative BMA algorithm to two cancer datasets: breast cancer and diffuse large B-cell lymphoma (DLBCL) data. On the breast cancer data, the algorithm selected a total of 15 predictor genes across 84 contending models from the training data. The maximum likelihood estimates of the selected genes and the posterior probabilities of the selected models from the training data were used to divide patients in the test (or validation) dataset into high- and low-risk categories. Using the genes and models determined from the training data, we assigned patients from the test data into highly distinct risk groups (as indicated by a p-value of 7.26e-05 from the log-rank test). Moreover, we achieved comparable results using only the 5 top selected genes with 100% posterior probabilities. On the DLBCL data, our iterative BMA procedure selected a total of 25 genes across 3 contending models from the training data. Once again, we assigned the patients in the validation set to significantly distinct risk groups (p-value = 0.00139).

Conclusion: The strength of the iterative BMA algorithm for survival analysis lies in its ability to account for model uncertainty. The results from this study demonstrate that our procedure selects a small number of genes while eclipsing other methods in predictive performance, making it a highly accurate and cost-effective prognostic tool in the clinical setting.

Show MeSH
Related in: MedlinePlus