I have been trying to run the batch effect correction for a couple of weeks now, but the same error persists. The same error message appears even when I try to run it on the example data - “Error: Failed to read in the CSV file. Possible causes of error (last one being the most relevant):”. Something similar has been commented on in these two previous topics in the Forum: Batch Correction issues - even examples not eligible / Batch Effect Correction Error Formatting Troubleshooting
Additionally, Volcano Plot analysis/visualization is not functional today.
I would appreciate it if these two issues could be solved within the shortest delay, as we need to run them before the submission of a master’s thesis project.
I am experiencing the same issue with the Volcano Plot.
Nevertheless, if you need it in a hurry, the Fold Change analysis and Welch’s t-test run successfully for me. You can combine these data to build your own Volcano plot. That is what I am doing
Good luck with your thesis!
To the Metaboanlyst team, thanks for your work! Hope you are able to fix it soon!
I did not really need to plot it, just the info. The volcano is nothing else than
x = log2(Fold change)
y = -log10(pvalue)
This information is:
x: in the fold_change.csv from the Fold Change analysis (the first column is the raw fold change, the second is already lith log2 transformation)
y: in the t_test.csv from the t test section (first column is the t statistic, second column is the raw p.value, third is the -log10 of the raw p.value, but take into account that if you applied the False Discovery Rate, that is the last column, therefore this is the one you need to check for significance < 0.05. Also take into account that if you do not have homocedastic data you should select Welch or simmilar in the options before submitting.
That is where you can get the info to build it, then all the colouring will depend where will you graph it (excel, R, python…) If you don’t know about this, I always find usefull asking the AI (I have not tried asking the AI to build it as I am not sure to how extent I trust them… but I have asked them to help me or to teach me how to do it, you need patience but they are usually good)
