Using recursive partitioning analysis to evaluate compound selection methods.

S Stanley Young; Douglas M Hawkins

doi:10.1385/1-59259-802-1:317

Home/Archive/PubMed

PubMed · 15141119

Using recursive partitioning analysis to evaluate compound selection methods.

S Stanley Young·Douglas M Hawkins

View original source ↗Export metadata ↓

Abstract

The design and analysis of a screening set for high throughput screening is complex. We examine three statistical strategies for compound selection, random, clustering, and space-filling. We examine two types of chemical descriptors, BCUTs and principal components of Dragon Constitutional descriptors. Based on the predictive power of multiple tree recursive partitioning, we reached the following tentative conclusions. Random designs appear to be as good as clustering and space-filling designs. For analysis, BCUTs appear to be better than principal components scores based upon Constitutional Descriptors. We confirm previous results that model-based selection of compounds can lead to improved screening hit rates.

Explore related subjects

Decision Trees Drug Design Evaluation Study Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

S Stanley Young, Douglas M Hawkins. 2004. Using recursive partitioning analysis to evaluate compound selection methods.. https://doi.org/10.1385/1-59259-802-1%3A317

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2005

EAU guidelines on urological trauma.

PURPOSE: To determine the optimal evaluation and management of genitourinary (renal, ureteral, bladder, urethral and genital) injuries by review of the world's literature on the subject. METHODS: A consensus committee convened by the Health Care Office of the European Association of Urology (EAU) to summarize the literature concerning the diagnosis and treatment of genitourinary trauma. RESULTS: Findings of 350 citations are reviewed. CONCLUSIONS: The genitourinary trauma literature still relies heavily on expert opinion and single-institution retrospective series. Future prospective trials of the most significant issues, when possible, might improve the quality of evidence that dictates practitioner behavior.

Decision Trees↗

PubMed2004

Safety issues of botanicals and botanical preparations in functional foods.

Although botanicals have played a role in the marketing of health products for ages, there is an increased interest today due to their perceived health benefits. Not only do consumers increasingly take charge of their health, but the scientific information and understanding of the beneficial health effects of bioactive substances in food, functional foods and food supplements have improved. Increasing use of these products has also led to concerns about their actual safety. Recorded cases of intoxications have triggered such concerns. The safety assessment of these substances is complicated by, amongst others, the variability of composition. Furthermore, consumption of such functional products is expected to produce physiological effects, which may lead to low margins of safety as the margin between exposure of such products and the safe level of intake are likely to be small. The safety assessment of botanicals and botanical preparations in food and food supplement should at least involve: the characterisation and quality of the material, its quality control; the intended use and consequent exposure; history of use and exposure; product comparison(s); toxicological information gathering; Risk characterisation/safety assessment. As a guidance tool, a decision tree approach is proposed to assist in determining the extent of data requirements based on the nature of the such product. This guidance tool in safety assessment was developed by an expert group of the International Life Sciences Institute (ILSI), European Branch, and is currently in press. In this paper a summarised version of this tool is presented.

Decision Trees↗

PubMed2004

Novel analytical methods applied to type 1 diabetes genome-scan data.

Complex traits like type 1 diabetes mellitus (T1DM) are generally taken to be under the influence of multiple genes interacting with each other to confer disease susceptibility and/or protection. Although novel methods are being developed, analyses of whole-genome scans are most often performed with multipoint methods that work under the assumption that multiple trait loci are unrelated to each other; that is, most models specify the effect of only one locus at a time. We have applied a novel approach, which includes decision-tree construction and artificial neural networks, to the analysis of T1DM genome-scan data. We demonstrate that this approach (1) allows identification of all major susceptibility loci identified by nonparametric linkage analysis, (2) identifies a number of novel regions as well as combinations of markers with predictive value for T1DM, and (3) may be useful in characterizing markers in linkage disequilibrium with protective-gene variants. Furthermore, the approach outlined here permits combined analyses of genetic-marker data and information on environmental and clinical covariates.

Decision Trees↗