Decision tree methods in pharmaceutical research.

Paul E Blower; Kevin P Cross

doi:10.2174/156802606775193301

Home/Archive/PubMed

PubMed · 16454756

Decision tree methods in pharmaceutical research.

Paul E Blower·Kevin P Cross

View original source ↗Export metadata ↓

Abstract

Decision trees are among the most popular of the new statistical learning methods being used in the pharmaceutical industry for predicting quantitative structure-activity relationships. This article reviews applications of decision trees in drug discovery research and extensions to the basic algorithm using hybrid or ensemble methods that improve prediction accuracy.

Explore related subjects

Chemistry, Pharmaceutical Decision Trees Drug Design Humans Quantitative Structure-Activity Relationship Journal Article Review

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Paul E Blower, Kevin P Cross. 2006. Decision tree methods in pharmaceutical research.. https://doi.org/10.2174/156802606775193301

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

KEEP EXPLORING

Related citations

PubMed2007

Alternative statistical parameter for high-throughput screening assay quality assessment.

High-throughput screening is an essential process in drug discovery. The ability to identify true active compounds depends on the high quality of assays and proper analysis of data. The Z factor, presented by Zhang et al. in 1999, provides an easy and useful summary of assay quality and has been a widely accepted standard. However, as data analysis has undergone much improvement recently, the assessment of assay quality has not evolved in parallel. In this article, the authors study the implications of Z factor values under different conditions and link the Z factor with the power of discovering true active compounds. They discuss the different interpretations of Z factor depending on error distributions and advocate direct analysis of power as assay quality assessment. They also propose that in estimating assay quality parameters, adjustments in data analysis should be taken into account. Studying the power of identifying true "hits" gives a more direct interpretation of assay quality and may provide guidance in assay optimization on some occasions.

Chemistry, Pharmaceutical↗

PubMed2007

Nonaqueous capillary electrophoresis in pharmaceutical analysis.

This review presents different solvents and electrolytes commonly used as BGEs in NACE for the analysis of pharmaceutical compounds. Most NACE applications carried out since 1998 for the analysis of compounds of pharmaceutical interest are presented in four tables: (i) analysis of drugs and related substances, (ii) analysis of chiral substances, (iii) analysis of phytochemical extracts and (iv) analysis of drugs in biological fluids. These selected examples are used to illustrate the interest in NACE versus conventional aqueous CE.

Chemistry, Pharmaceutical↗

PubMed2007

Recent progress in capillary ITP.

ITP has been attracting constant attention for many years due to its principal capability to concentrate trace analytes by several orders of magnitude. In the current capillary format, it is able to concentrate trace analytes diluted to several microliters of an original sample into concentrated zones having volumes in the range of picoliters. Due to this reason, ITP holds an important position in many current multistage and multidimensional separation schemes. This article links up previous reviews on the topic and summarizes the progress of analytical capillary ITP since 2002. Almost 100 papers are reviewed that include methodological novelties, instrumental aspects, and analytical applications. Papers using ITP and/or isotachophoretic principles as part of multistage and/or multidimensional separation schemes are also included.

Chemistry, Pharmaceutical↗