RESUMO
BACKGROUND: Data from patients with rare diseases is often produced using different platforms and probe sets because patients are widely distributed in space and time. Aggregating such data requires a method of normalization that makes patient records comparable. RESULTS: This paper proposed DBNorm, implemented as an R package, is an algorithm that normalizes arbitrarily distributed data to a common, comparable form. Specifically, DBNorm merges data distributions by fitting functions to each of them, and using the probability of each element drawn from the fitted distribution to merge it into a global distribution. DBNorm contains state-of-the-art fitting functions including Polynomial, Fourier and Gaussian distributions, and also allows users to define their own fitting functions if required. CONCLUSIONS: The performance of DBNorm is compared with z-score, average difference, quantile normalization and ComBat on a set of datasets, including several that are publically available. The performance of these normalization methods are compared using statistics, visualization, and classification when class labels are known based on a number of self-generated and public microarray datasets. The experimental results show that DBNorm achieves better normalization results than conventional methods. Finally, the approach has the potential to be applicable outside bioinformatics analysis.
Assuntos
Análise de Sequência com Séries de Oligonucleotídeos/métodos , Software , Área Sob a Curva , Regulação Neoplásica da Expressão Gênica , Humanos , Distribuição Normal , Leucemia-Linfoma Linfoblástico de Células Precursoras/genética , Leucemia-Linfoma Linfoblástico de Células Precursoras/patologia , Análise de Componente Principal , Curva ROC , Interface Usuário-ComputadorRESUMO
Datasets of offender attributes, both pre-custody and in-custody, were provided by the Correctional Service of Canada with the goal of exploring whether Security Threat Group (STG) offenders (informally, gang members of various kinds) differ in any systematic way from other offenders. For pre-custody attributes, we show that the entire offender population varies along two almost independent axes, one associated with affinity for violence, and the other with affinity for substance abuse. Within this structure, STG offenders are characteristically less extreme, in either direction, than the general offender population. For approximately two dozen attributes, STG offenders, as a group, tend to have higher values; for a few, they tend to have lower values. For in-custody attributes, the entire offender population forms a triangular structure whose vertices represent: passivity; violence and troublemaking; and involvement in programs leading to partial release. The differences between the STG offender population and the general offender population are small. An offender who is placed at the high end of the propensity for violence axis and/or the high end of the substance abuse axis based on pre-custody attributes is much more likely to be involved in incidents, grievances, and violence while in custody. This may have implications for risk stratification of incoming offenders.