Dataset Information

K-Nearest neighbor models for microarray gene expression analysis and clinical outcome prediction.

ABSTRACT: In the clinical application of genomic data analysis and modeling, a number of factors contribute to the performance of disease classification and clinical outcome prediction. This study focuses on the k-nearest neighbor (KNN) modeling strategy and its clinical use. Although KNN is simple and clinically appealing, large performance variations were found among experienced data analysis teams in the MicroArray Quality Control Phase II (MAQC-II) project. For clinical end points and controls from breast cancer, neuroblastoma and multiple myeloma, we systematically generated 463,320 KNN models by varying feature ranking method, number of features, distance metric, number of neighbors, vote weighting and decision threshold. We identified factors that contribute to the MAQC-II project performance variation, and validated a KNN data analysis protocol using a newly generated clinical data set with 478 neuroblastoma patients. We interpreted the biological and practical significance of the derived KNN models, and compared their performance with existing clinical factors.

SUBMITTER: Parry RM

PROVIDER: S-EPMC2920072 | biostudies-literature | 2010 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

k-Nearest neighbor models for microarray gene expression analysis and clinical outcome prediction.

Parry R M RM Jones W W Stokes T H TH Phan J H JH Moffitt R A RA Fang H H Shi L L Oberthuer A A Fischer M M Tong W W Wang M D MD

The pharmacogenomics journal 20100801 4

In the clinical application of genomic data analysis and modeling, a number of factors contribute to the performance of disease classification and clinical outcome prediction. This study focuses on the k-nearest neighbor (KNN) modeling strategy and its clinical use. Although KNN is simple and clinically appealing, large performance variations were found among experienced data analysis teams in the MicroArray Quality Control Phase II (MAQC-II) project. For clinical end points and controls from br ...[more]

PMID: 20676068

Similar Datasets

Project description:Many humans live in large, complex political centers, composed of multi-scalar communities including neighborhoods and districts. Both today and in the past, neighborhoods form a fundamental part of cities and are defined by their spatial, architectural, and material elements. Neighborhoods existed in ancient centers of various scales, and multiple methods have been employed to identify ancient neighborhoods in archaeological contexts. However, the use of different methods for neighborhood identification within the same spatiotemporal setting results in challenges for comparisons within and between ancient societies. Here, we focus on using a single method-combining Average Nearest Neighbor (ANN) and Kernel Density (KD) analyses of household groups-to identify potential neighborhoods based on clusters of households at 23 ancient centers across the Maya Lowlands. While a one-size-fits all model does not work for neighborhood identification everywhere, the ANN/KD method provides quantifiable data on the clustering of ancient households, which can be linked to environmental zones and urban scale. We found that centers in river valleys exhibited greater household clustering compared to centers in upland and escarpment environments. Settlement patterns on flat plains were more dispersed, with little discrete spatial clustering of households. Furthermore, we categorized the ancient Maya centers into discrete urban scales, finding that larger centers had greater variation in household spacing compared to medium-sized and smaller centers. Many larger political centers possess heterogeneity in household clustering between their civic-ceremonial cores, immediate hinterlands, and far peripheries. Smaller centers exhibit greater household clustering compared to larger ones. This paper quantitatively assesses household clustering among nearly two dozen centers across the Maya Lowlands, linking environment and urban scale to settlement patterns. The findings are applicable to ancient societies and modern cities alike; understanding how humans form multi-scalar social groupings, such as neighborhoods, is fundamental to human experience and social organization.

Dataset Information

K-Nearest neighbor models for microarray gene expression analysis and clinical outcome prediction.

Publications

k-Nearest neighbor models for microarray gene expression analysis and clinical outcome prediction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets