On 2017 Nov 12, Jonathan Eisen commented:

I believe the claims regarding "beneficial" organisms in this paper are inaccurate and misleading. For example, consider statement in the abstract:

"While the vast majority of microbial species classified were beneficial"

No evidence is presented anywhere in the paper that the microbes they identify via sequence analysis are beneficial in any way. I engaged in a Twitter discussion with the senior author of this paper, Chris Mason, about this topic where we discussed my concerns. Details of this discussion are here: https://phylogenomics.blogspot.com/2017/11/flaws-in-prediction-of-presence-of.html

As far as I can tell from this discussion and from the paper, the authors considered all organisms that were not specifically assigned to be a putative pathogen to be beneficial.

This is simply a flawed approach.

First, just because they are not assigned to be putative pathogenic does not mean that they are not pathogenic.

Second, this ignores the possibility that microbes could have no effect. That is, some could be parasitic, some could be beneficial, and some could have no effect - what was the no effect category ignored?

Third, the DNA could be coming from dead organisms that presumably would not have any significant effects.

Overall, I believe this paper's claims regarding "beneficial" microbes are inappropriate and misleading.

On 2017 Nov 12, Jonathan Eisen commented:

I believe the claims regarding "beneficial" organisms in this paper are inaccurate and misleading. For example, consider statement in the abstract:

"While the vast majority of microbial species classified were beneficial"

No evidence is presented anywhere in the paper that the microbes they identify via sequence analysis are beneficial in any way. I engaged in a Twitter discussion with the senior author of this paper, Chris Mason, about this topic where we discussed my concerns. Details of this discussion are here: https://phylogenomics.blogspot.com/2017/11/flaws-in-prediction-of-presence-of.html

As far as I can tell from this discussion and from the paper, the authors considered all organisms that were not specifically assigned to be a putative pathogen to be beneficial.

This is simply a flawed approach.

First, just because they are not assigned to be putative pathogenic does not mean that they are not pathogenic.

Second, this ignores the possibility that microbes could have no effect. That is, some could be parasitic, some could be beneficial, and some could have no effect - what was the no effect category ignored?

Third, the DNA could be coming from dead organisms that presumably would not have any significant effects.

Overall, I believe this paper's claims regarding "beneficial" microbes are inappropriate and misleading.