What to do with ALDEx2 associations where p < 0.05, but the 95% confidence intervals bridge zero

Question

[Sam]

Hi everyone,

I am using ALDEx2 to analyse the differential abundance of genera between two groups; however, the associations it calls significant (BH-corrected p < 0.05) all have 95% confidence intervals bridging zero. My questions are: Is this a known issue with ALDEx2? How does it calculate its confidence intervals? and in this case which should I trust more, the confidence intervals or the p-values?

Many thanks