minor comments for revision letter

Author: GuyBaele

revision_letter_new.tex

@@ -129,22 +129,22 @@
 
 \begin{reply}
 We thank the Editor-in-Chief and agree that the Reviewers' comments have helped us improve our manuscript. 
-As the referees shared some of the same concerns, we address those in general comments.
+As the referees shared some of the same concerns, we first address those in general comments.
 \end{reply}
 
 
 \textbf{General comment \#1: New data}
 
 \begin{reply}
 A major criticism of our original submission was that we did not use all of the publicly available data.
-To address this we conducted a through search of GenBank, now described in the Methods section of the paper:
+To address this we conducted a thorough search of GenBank, now described in the Methods section of the paper:
 \begin{quote}
 We retrieved all FMDV nucleotide sequences available from GenBank~\citep{Benson2013} from the National Center for Biotechnology Information (NCBI, \url{ http://www.ncbi.nlm.nih.gov/}) with more than $600$ bp.
 This first step yielded $6, 907$ sequences which were then filtered to exclude all sequences that did not include the 1D (VP1) gene, resulting in $4, 507$ sequences being kept.
 We then filtered for sequences from serotypes A and O, yielding $1051$ and $2350$ sequences, respectively.
 Next, we excluded sequences that had been extensively passaged in cell culture and selected all sequences from South America (Argentina, Bolivia, Brazil, Colombia, Ecuador, Paraguay, Peru, Uruguay, Venezuela) for which information on country and year of isolation was available.
 \end{quote}
-This procedure lead to $53$ additional sequences for serotype A and $43$ sequences for serotype O. 
+This procedure lead to $53$ additional sequences for serotype A and $43$ sequences for serotype O, compared to our original submission. 
 \end{reply}
 
 
@@ -155,7 +155,7 @@
 We address this important concern by employing analytical methods that explicitly accommodate the possibility of sampling bias.
 
 On the temporal side, we employ the modelling framework of~\cite{Karcher2020} to fit various coalescent-based models that specify the dependence of the sampling process on the population size or other temporal factors, while also accounting for phylogenetic uncertainty.
-Using with marginal likelihoods, we can compare these models and infer not only whether there is significant temporal bias but also possible explanatory factors.
+Using marginal likelihood estimation, we can compare these models to one another and infer not only whether there is significant temporal bias but also possible explanatory factors.
 
 To account for spatial bias, we employed a general linear model (GLM)~\citep{Lemey2014,Dudas2017} that allows for several predictors of spatial spread to be included simultaneously.
 This has the benefit of allowing one to construct predictors that account for possible sampling bias, such as the difference in numbers of sequences between locations and also the numbers of sequences at both origin and destination.
@@ -204,8 +204,9 @@
 \begin{reply}
 The fact that we use VP1 sequences is explicit in the methodology and all of our data and code are publicly available.
 In any case, we have now added a whole section at the end of the paper called ``Limitations of this study'' to address concerns of partial versus full genomes, which we agree is an important caveat.
-We have also included a citation of Valdazo-Gonzalez et al., (2012) in our discussion of full genomes~\textit{versus} partial sequences.
+We have also included a citation of Valdazo-Gonzalez et al. (2012) in our discussion of full genomes~\textit{versus} partial sequences.
 \end{reply}
+%GB: should we elaborate (here) more on the reason(s) behind our choice for VP1?
 
 4) The nature of the samples used for the virus sequence determination is also not indicated, have some been extensively passaged in cell culture? 
 This information should be available using the accession numbers of the published sequences and clearly can influence the outcome of sequence comparisons.
@@ -267,9 +268,9 @@
 \begin{reply}
 We thank the reviewer for such a through assessment of our work.
 It seems the reviewer's main concerns were (i) incomplete sampling of available data; (ii) sampling bias (even in face of all available data) and (iii) the level of generality of the results in face of (i) and (ii).
-We have now re-done the data collection and improved the data sets we analyse.
+We have now re-done the data collection and improved the data sets we analyse. %GB: improved in what way?
 We also employ better models that account for sampling bias both in time and space.
-For more please see responses below and also General Comments \#1 and \#2.
+For more elaborate information, please see tje responses below and also General Comments \#1 and \#2 above.
 \end{reply}
 
 Problems of the study:
@@ -298,10 +299,10 @@
 The only part which might be more realistic is the FMD transboundary movements within the ``countries triangle'' of Colombia, Ecuador and Venezuela, that could sounds more like from Argentina-Venezuela-Colombia-Ecuador, even though you have the effect of sampling bias that needs to be discussed.
 
 \begin{reply}
-We share the reviewers concerns that sampling bias might be an important factor in our analyses.
+We share the reviewer's concerns that sampling bias might be an important factor in our analyses.
 This has been addressed using state-of-the-art phylodynamic methods which accommodate both temporal and spatial sampling bias.
 While not a panacea, these methods are the best one can do in face of incomplete and potentially biased data.
-For more, see General Comment \#2.
+For more information, please see General Comment \#2.
 \end{reply}
 
 - Although you presented some data on sampling bias in your Supplementary Text (but this might be not satisfactory enough given the problem in the dataset), there is a real problem of sampling bias and this need to be addressed in your main text as well. 
@@ -323,15 +324,16 @@
 This information should be included either as a table in the main text or as a S3.
 
 \begin{reply}
-This list is now available from the PDF of the Supplementary Material.
+This list is now available in Supplementary Material.
 \end{reply}
 
 - This paper has been already published as an arXiv (http://arxiv.org/abs/1505.01105) which has been submitted the 5th of May and updated the 2nd of June (I received this review the 12th of June). 
 Although it is common for theoretical maths, physics and computer sciences studies to be published as arXiv before being properly peer-reviewed, this is not the case with study dealing with topics as in this case. 
 Since the study needs a substantial review and re-analysis of the data, I would suggest to withdraw your submission to arXiv
 
 \begin{reply}
-This comment makes no sense in light of the journal's policy.
+Our submission of the initial version of our manuscript to arXiv does not pose any conflict with the journal's policy.
+Additionally, arXiv and bioRxiv are well known at this point to host pre-prints that have not yet been peer-reviewed, a practice that is commonly accepted in our field of research (as can be seen from the many pre-prints circulating without peer review on SARS-CoV-2).
 \end{reply}
 
 Major Comments:
@@ -409,7 +411,7 @@
 Considering that you have a bias in your sequences for type A, you need to really consider with caution your results and discuss more about the impact it might cause.
 
 \begin{reply}
-The question of bias has been addressed more thoroughly (see General Comment \#2).
+The question of bias has now been addressed more thoroughly (see General Comment \#2).
 \end{reply}
 
 - Page 5 Line 16: ``Similar to what was found for Venezuela''. 
@@ -550,7 +552,7 @@
 - Page 2 Line 17: References 4 and 14 are duplicates.
 
 \begin{reply}
-We thank the reviewer for catching this blunder.
+We thank the reviewer for catching this.
 \end{reply}
 
 - Page 2 Line 17: Use Di Nardo et al. [12].
@@ -572,7 +574,7 @@
 Or you are only referring to South America?
 
 \begin{reply}
-South America is a continent.
+We are referring to South America, which is a continent so both statements would be correct.
 \end{reply}
 
 - Page 3 Line 48: ``..we employ an asymmetric''. 
@@ -596,7 +598,7 @@
 I think it would be better to say ``we obtained data from'' because maybe you have not been in the field collecting data.
 
 \begin{reply}
-This seems like a menial distinction.
+This seems like a menial distinction. %GB: I would just make the reviewer happy here, and say: corrected.
 \end{reply}
 
 - Page 7 Line 22: I would rather use: ``see Figure 3 in [6]'' or ``see FMD historical outbreak data in [6]''.
@@ -611,15 +613,15 @@
 I suggest using viral diversity throughout the paper.
 
 \begin{reply}
-
+%GB: simply replace in the text? Or avoid the use of Ne altogether and simply mention 'effective population size'?
 \end{reply}
 
 - Page 9 Line 13: It seems you used BEAST 1.7.5 - or even and older 1.7.2 version - (from your web available .xml) to perform the analyses, although a more recent version is available (1.8.2). 
 Is the latest version more robust and efficient in the results generated? 
 Have you re-analysed your data using the latest version and producing similar results?
 
 \begin{reply}
-All analyses have been conducted with the latest stable version of BEAST (v1.10). 
+All analyses have now been conducted with the latest stable version of BEAST (v1.10). 
 \end{reply}
 
 - You have a type O sequence from Peru (GenBank Accession No. HQ695844.1) you say collected in 2004, whilst in your previous paper on Ecuador is defined as 1994.