I noticed that the production software created an unfortunate error in the reference list. It split my reference # 32 into two items, generated # 33 for the second part of this reference, and as a result created a one item shift for all references from # 33 to # 70. I reported it to the PLoS production group and hope that they will fix this error sometime. However, for now reference # 33 in the text corresponds to the reference # 34 in the reference list, #34 in the text to #35 in the reference list, .... #70 in the text corresponds to #71 in the list. Sorry for inconvenience, I realize that it must be quite unpleasant. That brings me to the second part of my comment: I think it is unfortunate that PLoS One does not provide authors with the access to the galley proofs prior to publishing the paper. I just had another paper (on completely different topic) accepted by Nature. Scientific Reports ( another major online open access journal). They provide authors with galley proofs, which helps to avoid errors like the above one. Maybe PLoS One could provide us with galleys as well. Other than that my experience interacting with PLoS One was very pleasant: insightful comments from reviewers and editor, helpful staff.