Statistical Analysis [metadata table] - Time series + one condition - ANOVA Problem

I am VERY new to this software and to metabolomic analysis. I went to a workshop at the University of Illinois and got the basics to start analyzing data from one of my study. The study was a prospective, randomized, placebo control, double-blind clinical study in diabetic dogs who received either lyophilized feces (fecal microbiota transplantation) or placebo orally once daily for two months. I attached the metadata file and a sample of the LCMS data (with the first 25 metabolites) that I received from the University Metabolomic Core Facility. My data is balanced and I do not have any missing values. I followed tutorials, and matched my dataset formatting to the one that is given as an example for this analysis (cress_time.csv). I used log transformation and autoscaling to normalize the data. I chose time as a continuous variable (phenotype and subjects were set as categorical).

All works fine until I get to the ANOVA part where I get an error (see attached screenshots). If anyone has an idea of how to resolve it it’d be great.

Many thanks!!


LCMS_metabostat_metadata.csv (1.1 KB)
LCMS_metabostat_transposed.csv (17.7 KB)

The ANOVA for “time-series + one factor” is designed to evaluate the effect of the “one factor” across “time” . The “subject” should not be selected as one of the two metadata of interest, as it is already been accounted for in the “between subject” parameter.


Thanks for your reply.
When I choose phenotype and time I get the following error.
Any help would be greatly appreciated.

Also, I do not know why but when I have been attempting to redo the whole process the software does not let me change time to covariate so it is defaulted to be a categorical variable.



I tried your data - all worked as expected. I don’t see the issues you mentioned.

Thank you very much again for following up :slight_smile:
Does the ANOVA analysis has a limitation on the number of metabolites that are analyzed? I started to remove columns of metabolites (26 each time) and by process of elimination got to have the analysis working when I have metabolites up to column GS. Once I try to run it with one additional metabolite (i.e., up to column GT) I get the error which I posted below. I attach screenshots of both the successful analysis with the file containing GS columns and the unsuccessful one with the file containing GT columns. I also attach both CSV files.
Kindly let me know if there is a limitation on number of metabolites that can be analyzed at any given time.
Many thanks,
LCMS_metabostat_transposed GS.csv (141.1 KB)
LCMS_metabostat_transposed GT.csv (141.8 KB)

I think the answer to your question and alternative suggestions are both on the screenshot with bold fonts

This topic was automatically closed after 3 days. New replies are no longer allowed.