Dear Dr. Purkamo,
I can now inform you that both reviewers and the editor have evaluated carefully your manuscript: ‘The keystone species of Precambrian deep bedrock biosphere belong to Burkholderiales and Clostridiales’. Very little is known about the deep biosphere, and we are eager to understand its diversity and ecosystems functions. Thus, your study will be of interest to many readers of Biogeosciences. However, as you have seen from the interactive comments published on Biogeosciences Discussions major revisions were requested from both anonymous referees.

Both referees raised serious concerns regarding the reliability of metagenomic predictions especially for quantitative analysis of certain pathways. I agree with their comments that we should not overstate our findings and relate phylogeny to functions based on metagenomic predictions even if 16S rRNA similarities might be higher than 97%. Molecular tools like metagenomic and metatranscriptomic analyses are much more straightforward, but amounts of DNA and RNA present in these deep groundwater might be too low and/or the bioinformatics analyses too work-intensive. Nonetheless, I would like to ask you to tone down substantially this result part of your manuscript.

Please also omit the results of the HUMANn analysis.

I also do agree that you should change the title of your manuscript. There is no proof in your manuscript that you have identified the keystone species or that their removal from the groundwater would result in dramatic changes. It is well known that H2 is a key intermediate in the anaerobic degradation of organic matter and thus its production and consumption is key to all microbial communities in anoxic environments like sediments, wetlands, etc. What you present here are co-occurrence patterns and speculations about their metabolism.

Please also include the word groundwater or aquifers in your title, since you did not analyze the communities within the bedrock.

It would be helpful if you could also include some sentences about potential contaminations (presence of phototrophs) in the discussion section.

In general, I could see that most of the comments have been of use to you and I am looking forward with interest receiving your revised version. When submitting the revision, please also upload a separate file that carefully addresses, point-by-point, all issues raised by the referees. You should alos include a suitable rebuttal to any specific request for change that you have not made. State the line number of any revisions that are made.

Your interest in publishing in Biogeosciences is highly appreciated.

Kirsten Küsel

Dear Dr. Purkamo

Thanks for your changes. Thus, the revised version of your manuscript can be published in Biogeosciences without further delay.

Your interest in publishing in Biogeosciences is very much appreciated.

Sincerely,

Kirsten Küsel

(Associate Editor)