Hi @Yi_Xu ,
The problem of the first message is that, with the current parameters, you cannot run StrainPhlAn in your samples because, due to lack of markers, too many samples have been discarded. By default, StrainPhlAn needs that, at least, the 80% of the available markers of a species are present in, at least, 80% of the samples. Lowering the --sample_with_n_markers and --markers_in_n_samples will relax this thresholds. However, you cannot lower it down to 0, that will produce samples without markers and markers in no samples and thus an error in the execution.
Best,
Aitor