Hi, I am just learning to use MicrobiomeAnalyst in R and to check if I have done everything correctly I am using the online version. Now I see that my P-values for alpha diversity are slightly different when I use the filtered data. This is not the case with the original data. The filter criteria are the same. Could this be an installation problem? I Am using R 4.3.1 and installed MicrobiomeAnalystR last week. Thank you for your help!
Hello,
I have learn a little bit more about the problem…
If I upload my data online, exclude a sample from the analysis, edit the data using the default settings (filtering/normalisation) and then run an alpha diversity analysis, I get a p-value (e.g. 0.207). If I apply the data filter again and exclude the same sample, then use the same default settings for filtering and normalisation and then click on alpha diversity again, the p-value changes (e.g. 0.325) (the box plot also looks different). However, this p-value and boxplot then looks the same as when I do the analysis with R.
I have now observed this several times and also checked whether I clicked submit when removing the sample.
This dose not happen with the Original Data, here the p-value stays the same
mbSet<-Init.mbSetObj()
2. mbSet<-SetModuleType(mbSet, “mdp”)
3. mbSet<-ReadSampleTable(mbSet, “meta_upload.csv”);
4. mbSet<-Read16SAbundData(mbSet, “data_upload_anno.csv”,“text”,“QIIME”,“T”,“false”);
5. mbSet<-SanityCheckData(mbSet, “text”);
6. mbSet<-SanityCheckSampleData(mbSet);
7. mbSet<-SetMetaAttributes(mbSet, “1”)
8. mbSet<-PlotLibSizeView(mbSet, “norm_libsizes_0”,“png”);
9. mbSet<-CreatePhyloseqObj(mbSet, “text”,“QIIME”,“T” , “false”)
10. mbSet<-ApplyAbundanceFilter(mbSet, “prevalence”, 4, 0.2);
11. mbSet<-ApplyVarianceFilter(mbSet, “iqr”, 0.1);
12. smpl.nm.vec ← c(“1”,“13”,“16”,“19”,“22”,“25”,“28”,“31”,“34”,“4”,“7”,“11”,“14”,“17”,“2”,“20”,“23”,“26”,“29”,“32”,“35”,“5”,“8”,“12”,“15”,“18”,“21”,“24”,“27”,“3”,“30”,“33”,“36”,“6”,“9”)
13. mbSet<-UpdateSampleItems(mbSet);
14. mbSet<-PerformNormalization(mbSet, “none”, “colsum”, “none”, “true”);
15. mbSet<-PlotAlphaData(mbSet, “filt”,“alpha_diver_0”,“Chao1”,“group”,“OTU”, “default”, “png”);
16. mbSet<-PlotAlphaBoxData(mbSet, “alpha_diverbox_0”,“Chao1”,“group”,“default”, “png”);
17. mbSet<-PerformAlphaDiversityComp(mbSet, “tt”,“group”,“false”);
18. mbSet<-PlotAlphaData(mbSet, “filt”,“alpha_diver_1”,“Chao1”,“group”,“OTU”, “default”, “png”);
19. mbSet<-PlotAlphaBoxData(mbSet, “alpha_diverbox_1”,“Chao1”,“group”,“default”, “png”);
20. mbSet<-PerformAlphaDiversityComp(mbSet, “tt”,“group”,“true”);
21. smpl.nm.vec ← c(“1”,“13”,“16”,“19”,“22”,“25”,“28”,“31”,“34”,“4”,“7”,“11”,“14”,“17”,“2”,“20”,“23”,“26”,“29”,“32”,“35”,“5”,“8”,“12”,“15”,“18”,“21”,“24”,“27”,“3”,“30”,“33”,“36”,“6”,“9”)
22. mbSet<-UpdateSampleItems(mbSet);
23. mbSet<-ApplyAbundanceFilter(mbSet, “prevalence”, 4, 0.2);
24. mbSet<-ApplyVarianceFilter(mbSet, “iqr”, 0.1);
25. mbSet<-PerformNormalization(mbSet, “none”, “colsum”, “none”, “true”);
26. mbSet<-PlotAlphaData(mbSet, “filt”,“alpha_diver_2”,“Chao1”,“group”,“OTU”, “default”, “png”);
27. mbSet<-PlotAlphaBoxData(mbSet, “alpha_diverbox_2”,“Chao1”,“group”,“default”, “png”);
28. mbSet<-PerformAlphaDiversityComp(mbSet, “tt”,“group”,“true”);