Limits...
Statistical validation of megavariate effects in ASCA.

Vis DJ, Westerhuis JA, Smilde AK, van der Greef J - BMC Bioinformatics (2007)

Bottom Line: However, rigorous statistical validation of megavariate effects is still problematic because megavariate extensions of the classical F-test do not exist.If the observed effect is clearly different from this distribution the effect is deemed significant The permutation approach is studied using simulated data which gave successful results.In this metabolomics experiment the dosage and time-interaction effect were validated, both effects are significant.

View Article: PubMed Central - HTML - PubMed

Affiliation: BioSystems Data Analysis group, Swammerdam Institute for Life Science, University of Amsterdam, The Netherlands. science@danielvis.nl

ABSTRACT

Background: Innovative extensions of (M) ANOVA gain common ground for the analysis of designed metabolomics experiments. ASCA is such a multivariate analysis method; it has successfully estimated effects in megavariate metabolomics data from biological experiments. However, rigorous statistical validation of megavariate effects is still problematic because megavariate extensions of the classical F-test do not exist.

Methods: A permutation approach is used to validate megavariate effects observed with ASCA. By permuting the class labels of the underlying experimental design, a distribution of no-effect is calculated. If the observed effect is clearly different from this distribution the effect is deemed significant

Results: The permutation approach is studied using simulated data which gave successful results. It was then used on real-life metabolomics data set dealing with bromobenzene-dosed rats. In this metabolomics experiment the dosage and time-interaction effect were validated, both effects are significant. Histological screening of the treated rats' liver agrees with this finding.

Conclusion: The suggested procedure gives approximate p-values for testing effects underlying metabolomics data sets. Therefore, performing model validation is possible using the proposed procedure.

Show MeSH

Related in: MedlinePlus

Validation of the ASCA model for bromobenzene treated rats, validation of the dosage and the dosage-time interaction and the Xδ + Xτδ score plot. This experiment deals with the urine analysis of bromobenzene treated rats, the experimental design includes two types of controls and 3 dosage levels of the hepatotoxicant bromobenzene. The dosage and the interaction models are both significant as is clear from the reference distributions (p ≤ 0.0001). Because the dosage and the interaction models are significant they are superimposed and analyzed by SCA. The score plot of the SCA solution is shown. From this plot it is clear by visual inspection that the average dosage levels differ and that the interaction effect exists.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2211757&req=5

Figure 2: Validation of the ASCA model for bromobenzene treated rats, validation of the dosage and the dosage-time interaction and the Xδ + Xτδ score plot. This experiment deals with the urine analysis of bromobenzene treated rats, the experimental design includes two types of controls and 3 dosage levels of the hepatotoxicant bromobenzene. The dosage and the interaction models are both significant as is clear from the reference distributions (p ≤ 0.0001). Because the dosage and the interaction models are significant they are superimposed and analyzed by SCA. The score plot of the SCA solution is shown. From this plot it is clear by visual inspection that the average dosage levels differ and that the interaction effect exists.

Mentions: The main effects and the interaction effect of the 2-way ANOVA models were tested by the ASCA validation. Here the focus is on the factor dosage and dosage-time interaction. The models are significant, with a drug dose difference p ≤ 0.0001, SSQ = 3.181 (figure 2a) and dosage-time interaction p ≤ 0.0001, SSQ = 1.344 (figure 2b). The interaction significance was calculated on the residuals, thus after removing the time and dosage effect (equation 16). The not nested experimental design allows the use of a simple two-way ANOVA permutation scheme.


Statistical validation of megavariate effects in ASCA.

Vis DJ, Westerhuis JA, Smilde AK, van der Greef J - BMC Bioinformatics (2007)

Validation of the ASCA model for bromobenzene treated rats, validation of the dosage and the dosage-time interaction and the Xδ + Xτδ score plot. This experiment deals with the urine analysis of bromobenzene treated rats, the experimental design includes two types of controls and 3 dosage levels of the hepatotoxicant bromobenzene. The dosage and the interaction models are both significant as is clear from the reference distributions (p ≤ 0.0001). Because the dosage and the interaction models are significant they are superimposed and analyzed by SCA. The score plot of the SCA solution is shown. From this plot it is clear by visual inspection that the average dosage levels differ and that the interaction effect exists.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2211757&req=5

Figure 2: Validation of the ASCA model for bromobenzene treated rats, validation of the dosage and the dosage-time interaction and the Xδ + Xτδ score plot. This experiment deals with the urine analysis of bromobenzene treated rats, the experimental design includes two types of controls and 3 dosage levels of the hepatotoxicant bromobenzene. The dosage and the interaction models are both significant as is clear from the reference distributions (p ≤ 0.0001). Because the dosage and the interaction models are significant they are superimposed and analyzed by SCA. The score plot of the SCA solution is shown. From this plot it is clear by visual inspection that the average dosage levels differ and that the interaction effect exists.
Mentions: The main effects and the interaction effect of the 2-way ANOVA models were tested by the ASCA validation. Here the focus is on the factor dosage and dosage-time interaction. The models are significant, with a drug dose difference p ≤ 0.0001, SSQ = 3.181 (figure 2a) and dosage-time interaction p ≤ 0.0001, SSQ = 1.344 (figure 2b). The interaction significance was calculated on the residuals, thus after removing the time and dosage effect (equation 16). The not nested experimental design allows the use of a simple two-way ANOVA permutation scheme.

Bottom Line: However, rigorous statistical validation of megavariate effects is still problematic because megavariate extensions of the classical F-test do not exist.If the observed effect is clearly different from this distribution the effect is deemed significant The permutation approach is studied using simulated data which gave successful results.In this metabolomics experiment the dosage and time-interaction effect were validated, both effects are significant.

View Article: PubMed Central - HTML - PubMed

Affiliation: BioSystems Data Analysis group, Swammerdam Institute for Life Science, University of Amsterdam, The Netherlands. science@danielvis.nl

ABSTRACT

Background: Innovative extensions of (M) ANOVA gain common ground for the analysis of designed metabolomics experiments. ASCA is such a multivariate analysis method; it has successfully estimated effects in megavariate metabolomics data from biological experiments. However, rigorous statistical validation of megavariate effects is still problematic because megavariate extensions of the classical F-test do not exist.

Methods: A permutation approach is used to validate megavariate effects observed with ASCA. By permuting the class labels of the underlying experimental design, a distribution of no-effect is calculated. If the observed effect is clearly different from this distribution the effect is deemed significant

Results: The permutation approach is studied using simulated data which gave successful results. It was then used on real-life metabolomics data set dealing with bromobenzene-dosed rats. In this metabolomics experiment the dosage and time-interaction effect were validated, both effects are significant. Histological screening of the treated rats' liver agrees with this finding.

Conclusion: The suggested procedure gives approximate p-values for testing effects underlying metabolomics data sets. Therefore, performing model validation is possible using the proposed procedure.

Show MeSH
Related in: MedlinePlus