MetagenomSeq, perform anova or kruskall wallis with three groups ?
3
0
Entering edit mode
David ▴ 860
@david-3335
Last seen 3.6 years ago

Hi,

I'm trying to use metagenomSeq to perform something like a one-way ANOVA on 3
groups of samples. I have managed to succesfully normalize the data properly with metagenomeSeq but don´t know to run the anova or kruskall-w test with a metagenomeSeq object.

 

Thanks,

metagenomeseq anova metagenomics • 990 views
ADD COMMENT
0
Entering edit mode
@james-w-macdonald-5106
Last seen 15 hours ago
United States

I'm not sure metagenomeSeq is set up to do a Kruskal-Wallis. The closest it gets to doing non-parametric analysis is a permutation null distribution. Anyway, I think the conventional linear modeling is covered pretty well in section 4.2 of the vignette. Have you read that part? If so, is there something you don't understand?

ADD COMMENT
0
Entering edit mode
Hi David, The majority of the answer can be found in these responses: Moderated t-test in Anova limma - F-test for many condition One of the outputs of metagenomeSeq is a MLarrayM object. fit = fitZig(obj,mod) fit2 = eBayes(fit$fit) On fit2 you can then go ahead and use topTable from the limma package. On 2016-05-11 10:02, James W. MacDonald [bioc] wrote: > Activity on a post you are following on support.bioconductor.org [1] > > User James W. MacDonald [2] wrote Answer: MetagenomSeq, perform anova or kruskall wallis with three groups ? [3]: > > I'm not sure metagenomeSeq is set up to do a Kruskal-Wallis. The closest it gets to doing non-parametric analysis is a permutation null distribution. Anyway, I think the conventional linear modeling is covered pretty well in section 4.2 of the vignette. Have you read that part? If so, is there something you don't understand? > ------------------------- > > Post tags: metagenomeseq, anova, metagenomics > > You may reply via email or visit A: MetagenomSeq, perform anova or kruskall wallis with three groups ? Links: ------ [1] https://support.bioconductor.org [2] James W. MacDonald [3] A: MetagenomSeq, perform anova or kruskall wallis with three groups ?
ADD REPLY
0
Entering edit mode
David ▴ 860
@david-3335
Last seen 3.6 years ago

Ok i see,

I´m not familiar with limma.  I have tried the following:

fit = fitZig(obj,mod) 
fit2 = fit$fit 

topTable(fit2)

Error in topTable(fit2) : Need to run eBayes or treat first

fit2 <- ebayes(fit2)
topTable(fit2)
Error in topTable(fit3) : fit must be an MArrayLM object

Still don´t know how to get the overall pvalue and not the pvalue for all taxa.

Can i get the output of fitZig into a data.frame so i´m free to use aov() , tukeyHSD....

??????

Any idea.

> sessionInfo()
R version 3.2.4 (2016-03-10)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 14.04.4 LTS

locale:
 [1] LC_CTYPE=en_US.UTF-8          LC_NUMERIC=C                  LC_TIME=en_US.UTF-8           LC_COLLATE=en_US.UTF-8        LC_MONETARY=ca_AD.UTF-8       LC_MESSAGES=en_US.UTF-8      
 [7] LC_PAPER=ca_AD.UTF-8          LC_NAME=ca_AD.UTF-8           LC_ADDRESS=ca_AD.UTF-8        LC_TELEPHONE=ca_AD.UTF-8      LC_MEASUREMENT=ca_AD.UTF-8    LC_IDENTIFICATION=ca_AD.UTF-8

attached base packages:
[1] parallel  stats4    stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] xtable_1.8-2               biom_0.3.12                knitr_1.12.3               data.table_1.9.6           vegan_2.3-5                lattice_0.20-33           
 [7] permute_0.9-0              hexbin_1.27.1              vsn_3.38.0                 metagenomeSeq_1.12.1       glmnet_2.0-5               foreach_1.4.3             
[13] Matrix_1.2-6               limma_3.26.9               scales_0.4.0               gridExtra_2.2.1            gplots_3.0.1               xlsx_0.5.7                
[19] xlsxjars_0.6.1             rJava_0.9-8                RColorBrewer_1.1-2         DESeq2_1.10.1              RcppArmadillo_0.6.700.3.0  Rcpp_0.12.4               
[25] SummarizedExperiment_1.0.2 Biobase_2.30.0             GenomicRanges_1.22.4       GenomeInfoDb_1.6.3         IRanges_2.4.8              S4Vectors_0.8.11          
[31] BiocGenerics_0.16.1        phyloseq_1.14.0            psych_1.5.8                reshape2_1.4.1             ggplot2_2.1.0              plyr_1.8.3                
[37] stringr_1.0.0             

loaded via a namespace (and not attached):
 [1] nlme_3.1-127          bitops_1.0-6          matrixStats_0.50.2    tools_3.2.4           affyio_1.40.0         rpart_4.1-10          KernSmooth_2.23-15    Hmisc_3.17-4         
 [9] DBI_0.4               mgcv_1.8-12           colorspace_1.2-6      ade4_1.7-4            nnet_7.3-12           mnormt_1.5-4          compiler_3.2.4        preprocessCore_1.32.0
[17] chron_2.3-47          formatR_1.3           labeling_0.3          caTools_1.17.1        genefilter_1.52.1     affy_1.48.0           digest_0.6.9          foreign_0.8-66       
[25] rmarkdown_0.9.6       XVector_0.10.0        htmltools_0.3.5       highr_0.5.1           RSQLite_1.0.0         BiocInstaller_1.20.1  BiocParallel_1.4.3    gtools_3.5.0         
[33] acepack_1.3-3.3       magrittr_1.5          Formula_1.2-1         futile.logger_1.4.1   munsell_0.4.3         ape_3.4               yaml_2.1.13           stringi_1.0-1        
[41] MASS_7.3-45           RJSONIO_1.3-0         zlibbioc_1.16.0       grid_3.2.4            gdata_2.17.0          Biostrings_2.38.4     splines_3.2.4         multtest_2.26.0      
[49] annotate_1.48.0       locfit_1.5-9.1        igraph_1.0.1          geneplotter_1.48.0    codetools_0.2-14      futile.options_1.0.0  XML_3.98-1.4          evaluate_0.9         
[57] latticeExtra_0.6-28   lambda.r_1.1.7        gtable_0.2.0          survival_2.39-2       iterators_1.0.8       AnnotationDbi_1.32.3  cluster_2.0.4        
> 

 

ADD COMMENT
0
Entering edit mode
Instead of ebayes it should be eBayes. On 2016-05-12 08:30, David [bioc] wrote: > Activity on a post you are following on support.bioconductor.org [1] > > User David [2] wrote Answer: MetagenomSeq, perform anova or kruskall wallis with three groups ? [3]: > > Ok i see, > > I´m not familiar with limma. I have tried the following: > > fit = fitZig(obj,mod) > fit2 = fit$fit > > topTable(fit2) > > Error in topTable(fit2) : Need to run eBayes or treat first > > fit2 <- ebayes(fit2) > topTable(fit2) > Error in topTable(fit3) : fit must be an MArrayLM object > > Still don´t know how to get the overall pvalue and not the pvalue for all taxa. > > ?????? > > Any idea. > >> sessionInfo() > R version 3.2.4 (2016-03-10) > Platform: x86_64-pc-linux-gnu (64-bit) > Running under: Ubuntu 14.04.4 LTS > > locale: > [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 LC_MONETARY=ca_AD.UTF-8 LC_MESSAGES=en_US.UTF-8 > [7] LC_PAPER=ca_AD.UTF-8 LC_NAME=ca_AD.UTF-8 LC_ADDRESS=ca_AD.UTF-8 LC_TELEPHONE=ca_AD.UTF-8 LC_MEASUREMENT=ca_AD.UTF-8 LC_IDENTIFICATION=ca_AD.UTF-8 > > attached base packages: > [1] parallel stats4 stats graphics grDevices utils datasets methods base > > other attached packages: > [1] xtable_1.8-2 biom_0.3.12 knitr_1.12.3 data.table_1.9.6 vegan_2.3-5 lattice_0.20-33 > [7] permute_0.9-0 hexbin_1.27.1 vsn_3.38.0 metagenomeSeq_1.12.1 glmnet_2.0-5 foreach_1.4.3 > [13] Matrix_1.2-6 limma_3.26.9 scales_0.4.0 gridExtra_2.2.1 gplots_3.0.1 xlsx_0.5.7 > [19] xlsxjars_0.6.1 rJava_0.9-8 RColorBrewer_1.1-2 DESeq2_1.10.1 RcppArmadillo_0.6.700.3.0 Rcpp_0.12.4 > [25] SummarizedExperiment_1.0.2 Biobase_2.30.0 GenomicRanges_1.22.4 GenomeInfoDb_1.6.3 IRanges_2.4.8 S4Vectors_0.8.11 > [31] BiocGenerics_0.16.1 phyloseq_1.14.0 psych_1.5.8 reshape2_1.4.1 ggplot2_2.1.0 plyr_1.8.3 > [37] stringr_1.0.0 > > loaded via a namespace (and not attached): > [1] nlme_3.1-127 bitops_1.0-6 matrixStats_0.50.2 tools_3.2.4 affyio_1.40.0 rpart_4.1-10 KernSmooth_2.23-15 Hmisc_3.17-4 > [9] DBI_0.4 mgcv_1.8-12 colorspace_1.2-6 ade4_1.7-4 nnet_7.3-12 mnormt_1.5-4 compiler_3.2.4 preprocessCore_1.32.0 > [17] chron_2.3-47 formatR_1.3 labeling_0.3 caTools_1.17.1 genefilter_1.52.1 affy_1.48.0 digest_0.6.9 foreign_0.8-66 > [25] rmarkdown_0.9.6 XVector_0.10.0 htmltools_0.3.5 highr_0.5.1 RSQLite_1.0.0 BiocInstaller_1.20.1 BiocParallel_1.4.3 gtools_3.5.0 > [33] acepack_1.3-3.3 magrittr_1.5 Formula_1.2-1 futile.logger_1.4.1 munsell_0.4.3 ape_3.4 yaml_2.1.13 stringi_1.0-1 > [41] MASS_7.3-45 RJSONIO_1.3-0 zlibbioc_1.16.0 grid_3.2.4 gdata_2.17.0 Biostrings_2.38.4 splines_3.2.4 multtest_2.26.0 > [49] annotate_1.48.0 locfit_1.5-9.1 igraph_1.0.1 geneplotter_1.48.0 codetools_0.2-14 futile.options_1.0.0 XML_3.98-1.4 evaluate_0.9 > [57] latticeExtra_0.6-28 lambda.r_1.1.7 gtable_0.2.0 survival_2.39-2 iterators_1.0.8 AnnotationDbi_1.32.3 cluster_2.0.4 >> > > ------------------------- > > Post tags: metagenomeseq, anova, metagenomics > > You may reply via email or visit A: MetagenomSeq, perform anova or kruskall wallis with three groups ? Links: ------ [1] https://support.bioconductor.org [2] David [3] A: MetagenomSeq, perform anova or kruskall wallis with three groups ?
ADD REPLY
0
Entering edit mode
David ▴ 860
@david-3335
Last seen 3.6 years ago

Great !!!!

thanks

ADD COMMENT

Login before adding your answer.

Traffic: 402 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6