Dataset Information

AutoClass@IJM: a powerful tool for Bayesian classification of heterogeneous data in biology.

ABSTRACT: Recently, several theoretical and applied studies have shown that unsupervised Bayesian classification systems are of particular relevance for biological studies. However, these systems have not yet fully reached the biological community mainly because there are few freely available dedicated computer programs, and Bayesian clustering algorithms are known to be time consuming, which limits their usefulness when using personal computers. To overcome these limitations, we developed AutoClass@IJM, a computational resource with a web interface to AutoClass, a powerful unsupervised Bayesian classification system developed by the Ames Research Center at N.A.S.A. AutoClass has many powerful features with broad applications in biological sciences: (i) it determines the number of classes automatically, (ii) it allows the user to mix discrete and real valued data, (iii) it handles missing values. End users upload their data sets through our web interface; computations are then queued in our cluster server. When the clustering is completed, an URL to the results is sent back to the user by e-mail. AutoClass@IJM is freely available at: http://ytat2.ijm.univ-paris-diderot.fr/AutoclassAtIJM.html.

SUBMITTER: Achcar F

PROVIDER: S-EPMC2703914 | biostudies-literature | 2009 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

AutoClass@IJM: a powerful tool for Bayesian classification of heterogeneous data in biology.

Achcar Fiona F Camadro Jean-Michel JM Mestivier Denis D

Nucleic acids research 20090527 Web Server issue

Recently, several theoretical and applied studies have shown that unsupervised Bayesian classification systems are of particular relevance for biological studies. However, these systems have not yet fully reached the biological community mainly because there are few freely available dedicated computer programs, and Bayesian clustering algorithms are known to be time consuming, which limits their usefulness when using personal computers. To overcome these limitations, we developed AutoClass@IJM, ...[more]

PMID: 19474346

Similar Datasets

Project description:BackgroundCommonly, several traits are assessed in agronomic experiments to better understand the factors under study. However, it is also common to see that even when several traits are available, researchers opt to follow the easiest way by applying univariate analyses and post-hoc tests for mean comparison for each trait, which arouses the hypothesis that the benefits of a multi-trait framework analysis may have not been fully exploited in this area.ResultsIn this paper, we extended the theoretical foundations of the multi-trait genotype-ideotype distance index (MGIDI) to analyze multivariate data either in simple experiments (e.g., one-way layout with few treatments and traits) or complex experiments (e.g., with a factorial treatment structure). We proposed an optional weighting process that makes the ranking of treatments that stands out in traits with higher weights more likely. Its application is illustrated using (1) simulated data and (2) real data from a strawberry experiment that aims to select better factor combinations (namely, cultivar, transplant origin, and substrate mixture) based on the desired performance of 22 phenological, productive, physiological, and qualitative traits. Our results show that most of the strawberry traits are influenced by the cultivar, transplant origin, cultivation substrates, as well as by the interaction between cultivar and transplant origin. The MGIDI ranked the Albion cultivar originated from Imported transplants and the Camarosa cultivar originated from National transplants as the better factor combinations. The substrates with burned rice husk as the main component (70%) showed satisfactory physical proprieties, providing higher water use efficiency. The strengths and weakness view provided by the MGIDI revealed that looking for an ideal treatment should direct the efforts on increasing fruit production of Albion transplants from Imported origin. On the other hand, this treatment has strengths related to productive precocity, total soluble solids, and flesh firmness.ConclusionsOverall, this study opens the door to the use of MGIDI beyond the plant breeding context, providing a unique, practical, robust, and easy-to-handle multi-trait-based framework to analyze multivariate data. There is an exciting possibility for this to open up new avenues of research, mainly because using the MGIDI in future studies will dramatically reduce the number of tables/figures needed, serving as a powerful tool to guide researchers toward better treatment recommendations.

Dataset Information

AutoClass@IJM: a powerful tool for Bayesian classification of heterogeneous data in biology.

Publications

AutoClass@IJM: a powerful tool for Bayesian classification of heterogeneous data in biology.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets