Annette M Molinaro PhD
Visiting Associate Professor
Statistical genomics/genetics; Computational biology; Cancer; Prediction models; Risk models; Survival Analysis; partDSA; Recursive Partitioning
- Novel Statistical Methods for Predicting Clinical Outcomes and Assessing Variable Importance in the Presence of Competing Risks
Dr. Molinaro’s research interests are primarily focused on statistical genetics and computational biology, including prediction, survival analysis, classification, and causal inference with additional curiosities in cancer epidemiology and in the estimation of absolute risk in nested case-control studies. Her research has pertained to predicting clinical outcomes with high-dimensional explanatory variables, such as microarrays, and large-scale epidemiology studies. This has included an adaptation to Classification and Regression Trees (CART) for survival outcomes, the introduction of a novel data-adaptive algorithm that builds Boolean combinations of explanatory variables, and a non-parametric method for point estimation based on a nested case-control study design. In addition, she has worked with collaborators at the National Cancer Institute (NCI) on comparing cross-validation approaches to validating predictors in small sample sizes. Dr. Molinaro’s current focus is on exploring new algorithms for building predictors with high-dimensional data structures including genomic and proteomic data.
- Lostritto K, Strawderman R, and Molinaro AM. A Partitioning Deletion/Substitution/ Addition Algorithm for Creating Survival Risk Groups. In print, Biometrics. http://arxiv.org/abs/1101.4331.
- Molinaro AM, Carriero NJ, Bjornson R, Hartge P, Rothman N, and Chatterjee, N. Power of data mining methods to detect genetic associations and interactions. Human Heredity, 2011. 72:85-97 (DOI: 10.1159/000330579). [PMCID: PMC3222116].
- Tolles J, Bai Y, Baquero M, Harris LN, Rimm DL, and Molinaro AM. Optimal Tumor Sampling for Immunostaining of Biomarkers in Breast Carcinoma. Breast Cancer Research, 2011. May 18;13(3):R51. doi:10.1186/bcr2882. [PMCID: PMC3218938].
- Bai Y, Tolles J, Cheng H, Siddiqui S, Gopinath A, Pectasides E, Camp RL, Rimm DL, and Molinaro AM. Quantitative Assessment Shows Loss of Antigenic Epitopes as a Function of Time to Formalin Fixation. Laboratory Investigation, 2011. August; 91: 1253-1261 doi:10.1038/labinvest.2011.75. [PMCID: PMC3145004].
- Molinaro AM, Lostritto K, van der Laan MJ. partDSA: Deletion/Substitution/Additional Algorithm for Partitioning the Covariate Space in Prediction. Bioinformatics 26:1357-1363, 2010. [PMCID: PMC2865863].
- Kerlikowske, K./Molinaro, A. M./Gauthier M.*, Berman, H., Waldman, F., Bennington, J., Sanchex, H., Jimenez, C., Stewart, K., Chew, K., Ljung, B., and Tlsty, T. Biomarker Expression and Risk of Subsequent Tumors After Initial Ductal Carcinoma In Situ Diagnosis. J. Natl. Cancer Inst. 102: 627 637, 2010. (* Shared first authorship). [PMCID: PMC2864293].
- Pelizzola, M., Koga, Y., Urban, A.E., Krauthammer, M.,Weissman, S., Halaban, R., and Molinaro, A.M. MEDME: An experimental and analytical methodology for the estimation of DNA methylation levels based on microarray derived MeDIP-enrichment. Genome Research 18:1652–1659, 2008. [PMCID: PMC2556264].
- Hothorn, T., Buhlmann, P., Dudoit, S., Molinaro, A.M., and van der Laan, M.J. Survival Ensembles. Biostatistics 7: 355-373, 2006. [PMID: 16344280].
- Molinaro, A.M., Simon, R., and Pfeiffer, R.M. Prediction Error Estimation: A Comparison of Resampling Methods. Bioinformatics 21(15): 3301-3307, 2005. [PMID: 15905277].
- Molinaro, A.M., Dudoit, S., and van der Laan, M.J. Tree-based Multivariate Regression and Density Estimation With Right-censored Data. Journal of Multivariate Analysis 90(1): 154-177, 2004.