Closed lgeistlinger closed 1 month ago
I assume this will go on the Study form and, therefore, will not vary between Experiments. Please confirm. Also, should this be placed at the end of the form/content page? Or somewhere else?
Please let me know once the final options have been confirmed.
This will indeed go in the experiment form in the "Statistical analysis" section. It should be the first field, ie before "Statistical test", "Significance threshold", etc.
👍 for where @lgeistlinger said.
Will do. Should we include a tooltip? The other ones have it.
Yes: "Data transformation applied to microbial abundance measurements prior to differential abundance testing (if any)"
We implemented this, assuming that you can only have one value in this field. Please let me know if that is incorrect:
Looks good, thanks!
One minor modification - it should be "arcsine square-root" or "arcsine sqrt" rather than just "arcsine"
done
Can we add an additional option for "log transformation"? This is different from a central log ratio.
Sorry if there's a way I can add these myself but I couldn't find it.
Added. Note that autocomplete values can be modified here: https://bugsigdb.org/Help:Admin
One reviewer raised the following point: There was no indication whether the datasets had been subject to clr-transformation or other compositional data analysis methods. It would be nice to have this as a field or a checkbox since these data are all compositional, but most studies do not adjust for this. I it would be nice to know which are.
After discussion with @lwaldron, we decided that this would be best addressed via a new statistical analysis field called "Data transformation".