This manuscript describes an extensive analysis of the distributions of fold and superfamily protein structures in a large number of genomes. It is a comprehensive study that, in my opinion, contains information that will provide new insights into the biological basis of distributions and their evolutionary implications. The work is particularly valuable as a guide for target selection in the emerging area of structural proteomics, and as a benchmark analysis of the genome / protein structure landscape. For these reasons, I would like to see the work published in PSFG.
Because this analysis of domain structure distributions across many genomes is so extensive and complex, the mining of this data for biological understanding is definitely a challenging task, and will probably be done over time. In a sense, this paper functions to lead the reader to a more complete analysis that is available on the author's web site. This is also motivation to publish the work in PSFG, particularly if the published work extracts in a concise way the key novel information accessible through this website.
However, I do agree with the previous editorial opinion that the authors do not adequately describe the new insights they have gained from this analysis. Some interesting points are made regarding domain pair co-occurrences, the face that some superfamilies are unique to specific genomes, and the fact that some common superfamilies are absent from specific genomes. The analysis of these observations is more extensive in this work than in any previously published students that I am aware of. However, it sems that more effeot should be made to interpret some of these observations, either in terms of general trends, specific evolutionary implications, or with respect to specific biological processes. This is done in the curent version, but only to a very small extent.
In order to adequately address concerns raised in the previous reviews, and to do justice to the extensive and technically sound analysis which the authors have completed, I feel that there has to be more discussion of biological interpretations of the statistical observations presented in the paper. If this theme can be developed a bit, with at least some specific examples demonstrating biological insight, I could recommend the paper for publication.
I strongly encourage the authors to expand a bit more on the biological and/or evolutionary implications of their result, as I would very much like to see this work published in PSFG.
In addition to these primary general concerns, there are many features of the presentation which make it difficult to follow. Fig 1A provides little insight - it appears that ALL the boxes have dots in them, and the information in the shaded boxes is detailed in Fig 1B - I suggest removing Fig 1A - or at least labeling the axes and making clearer what its point it. Why is the organism called Aquifex in Fig 2A and Aaeo in Fig 2B? Fig 3. is misprinted. Fig 4 and Table II (sometimes called Table 2) would have much more valuable if additional specific examples of these cases were discussed, particularly with regard to biological insights. The "Insert Fig" notations throughout the text are inaccurate. Is it SCOP or Scop?