Katie’s research interests focus on analysis methods for large-scale data sets (e.g., microarray studies), with applications in cancer biology. Approaches are based on simulations and employ both parametric and non-parametric bootstrap sampling of microarray data. Methodological components include:

Developing a theoretical framework on which to assess the reproducibility and reliability of summary measures (e.g., sample means, sample correlations and clusters).

Employing this framework to compare the effectiveness of existing and novel clustering techniques (including two-way and iterative clustering) in terms of reproducibility.

Multiple hypothesis testing (e.g., identifying subsets of significantly differently expressed genes or genes significantly associated with an outcome).

Developing appropriate statistical techniques for analyzing longitudinal microarry data.

Developing statistical methods linking clinical outcomes and covariates to microarray data.

Determining reliable statistical methods for classification of tumors and patients based on clinical outcomes and covariates.

Comparing gene expression patterns between different types of experiments (e.g., "knock-out", cell line and patient data).

Developing techniques for visualizing patterns and relationships in high-dimensional data.

Publications (2001-2004):

For more recent publications, see docpollard.com.

M.J. van der Laan, S. Dudoit, K.S. Pollard (2004), Multiple Testing. Part III. Procedures for Control of the Generalized Family-Wise Error Rate and Proportion of False Positives, U.C. Berkeley Division of Biostatistics Working Paper Series. Working Paper 141, revised for Statistical Applications in Genetics and Molecular Biology.
(PDF (BEPRESS)) 

M.J. van der Laan, S. Dudoit, K.S. Pollard (2003), Multiple Testing. Part II. Step-Down Procedures for Control of the Family-Wise Error Rate, U.C. Berkeley Division of Biostatistics Working Paper Series. Working Paper 139, revised for Statistical Applications in Genetics and Molecular Biology.
(PDF (BEPRESS)) 

S. Dudoit, M.J. van der Laan, K.S. Pollard (2003), Multiple Testing. Part I. Single-Step Procedures for Control of General Type I Error Rates, U.C. Berkeley Division of Biostatistics Working Paper Series. Working Paper 138, revised for Statistical Applications in Genetics and Molecular Biology.
(PDF (BEPRESS)) 

K.S. Pollard, M.J. van der Laan (2003), Multiple testing for gene expression data: an investigation of null distributions with consequences for the permutation test, Proceedings of the 2003 International MultiConference in Computer Science and Engineering, METMBS'03 Conference, pp.3-9.
(PDF) 

K.S. Pollard, M.J. van der Laan (2002), Resampling-based Multiple Testing: Asymptotic Control of Type I Error and Applications to Gene Expression Data, revised for a special issue of Journal of Statistical Planning and Inference.
(PDF (BEPRESS)) 

K.S. Pollard, M.J. van der Laan (2002), New methods for identifying significant clusters in gene expression data, 2002 Proceedings of the American Statistical Association, Biometrics Section [CD-ROM].
(PDF) 

K.S. Pollard, M.J. van der Laan (2002), A method to identify significant clusters in gene expression data, Proceedings of SCI 2002: Vol. II, pp. 318-325.
(PDF (BEPRESS) | PostScript)

M.J. van der Laan, K.S. Pollard, J. Bryan (2003), A New Partitioning Around Medoids Algorithm, revised for Journal of Statistical Computation and Simulation: 73(8), pp.575-584.
(PDF (BEPRESS) | PostScript)

J. Bryan, K.S. Pollard, M.J. van der Laan (2002), Paired and Unpaired Comparison and Clustering with Gene Expression Data, revised for Statistica Sinica: 12(1), pp. 87-110.
(PDF (BEPRESS) | PostScript)

K.S. Pollard, M.J. van der Laan (2002), Statistical Inference for Simultaneous Clustering of Gene Expression Data, revised for Mathematical Biosciences: 176(1), pp. 99-121.
(PDF (BEPRESS) | PostScript)

M.J. van der Laan, K.S. Pollard (2001), Hybrid clustering of gene expression data with visualization and the bootstrap, revised for Journal of Statistical Planning and Inference, 2003, 117, pp. 275--303.
(PDF | PostScript)

Presentations(2001-2003):

Katie taught a session about the Bioconductor open source software project at the QB3 Microarray Course (August 15, 2003 at UC Santa Cruz). The course materials are available on the Bioconductor website.

Katie gave a talk entitled "Multiple testing for gene expression data: an investigation of null distributions with consequences for the permutation test" at METMBS'03: 2003 International Conference on Mathematics and Engineering Techniques in Medicine and Biological Sciences (June 23 - 26, 2003 in Las Vegas, NV). Slides are available in PDF format.

Slides from the seminar "Multiple testing for high dimensional biological data" (given March 20, 2003 at UC Berkeley) are available in PDF format.

Katie gave a talk entitled "Computationally intensive statistical methods for analysis of gene expression data" at the Center for Biomolecular Science and Engineering at UC Santa Cruz (January 14, 2003). Slides are available in PDF format.

Katie gave a talk entitled "A new method for finding significant clusters in gene expression data" at JSM 2002: The Joint Statistical Meetings (August 10-13, 2002 in New York, NY).

Katie contributed a talk at MCP 2002: The 3rd International Conference on Multiple Comparisons (August 5-7, 2002 in Bethesda, MD). The slides from her talk "Resampling-based methods for identification of significant subsets of genes in expression data" are available in PDF and PostScript formats.

Katie participated in the Student Paper Competition at the WNAR Annual Meeting (June 23-26, 2002 at UCLA). The slides from her talk "Methods for analysis of gene expression data with a right-censored outcome" are available in PDF and PostScript formats.

Katie was an invited speaker at the IPAM Functional Genomics Reunion Conference (June 17-21, 2002 at UCLA). The abstract and slides from her talk "Statistical inference for simultaneous clustering of gene expression data" are available on the conference website.

Slides from the seminar "A new method for finding significant clusters in gene expression data" (given February 14, 2002 at UC Berkeley) are available in PDF and PostScript formats.

Katie contributed a talk at the MSRI Workshop on Nonlinear Estimation and Classification (March 19-29, 2001), entitled "Computationally intensive statistical methods for microarray based drug discovery". The slides from this presentation are available in PDF and PostScript formats.

top