StrainPhlAn --print_clades_only lists clades not listed by MetaPhlAn2

[Monica Ticlla]

Hi,

I'm testing MetaPhlAn2 and StrainPhlAn with a small set of samples (n=3).

I' ve run StrainPhlAn on the sam files generated by MetaPhlAn2, 3 samples were included (background negative controls).

When using the option --print_clades_only, the list includes only 2 clades:

s__Human_endogenous_retrovirus_K
s__Propionibacterium_phage_PAD20

In the MetaPhlAn2 results the s__Human_endogenous_retrovirus_K is not present,

please see below the heatmap of the species-level abundance profile (all species included):

You can also see that  s__Propionibacterium_phage_PAD20 is not present in all the samples (according to MetaPhlAn2).

How can these differences be explained?

Thanks,

Monica

[aitor.blancomiguez]

Hi Monica,

The differences that you are seeing here are due the different ways MetaPhlAn2 and StrainPhlAn process the mapping results.

In some cases you will see species in MetaPhlAn2 that cannot be analysed by StrainPhlAn and vice versa. 

Moreover, analysing your heatmap, for most of the bugs you profiled, MetaPhlAn2 was only able to profile them until the genus level (the _unclassified species), so they will be out of the StrainPhlAn species analysis. 

I hope my answer has been helpful.

Best,

Aitor