Criteria for feature selection in MaAsLin3 summary plot

Downstream analysis and statistics
1

Hello,

I am using MaAsLin3 to analyze my microbiome data. In the output, I noticed that many features were listed for certain variables in the significant_results.tsv file. However, not all of these features appeared in the summary plot.

I am wondering what selection criteria MaAsLin3 uses to decide which microbiome features across variables are shown in the summary plot?

Thank you!

2

Hi,

All the associations in the significant_results.tsv are sorted by joint q-value, and then the top first_n (unique) features are selected from that list to be displayed. If a feature has multiple associations with different variables that are the most significant, it still only counts that feature once so that you always get first_n features in the plot.

Will