Referee 1's review:

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication, the manuscript has been revised in light of these comments and to address other editorial requirements.
**********

The authors present an analysis of the abstracts of the Annual Meeting of Neurosiences with the aim of providing an objective view of the current research trend in terms of demographics, research collaboration, funding resource and main research topics.

Abstract
There are several aspects form the abstract which are not very clear, one is related to the statement that the authors want to gain an objective insight if a given research field. Here in terms of the data collection used it is not very clear whether by using only a single journal this can be guaranteed. It seems also likely that the authors provide a view of the research trends published in a given journal (journal publishing policy) rather than covering a whole domain. The structure of the abstract could also be improved. It is also not very clear what the actual use of the provided analysis would be in practice. Here a key issue is also whether the presented strategy be applied to other research domains. It is also not very clear whether this particular topic is suitable/of interest for the readership of PLOS One.

Introduction
Here there authors do not mention several points, one is to provide a more in detail overview on previous work related to the presented research. The other is related to actually choosing the topic of neurosciences.
It is not clear why the authors used the abstracts only, and not the full text articles.
The authors mention the use of reputational indices but do not provide a corresponding reference. The authors should also mention previous work on co-author network extraction (there is at least one working online tool based on PubMed from the Gerstein lab) and also previous work related to citation network extracted from full text biomedical literature is missing. Regarding the monitoring of research trends, there are also several publications which could be mentioned here, e.g. related to research trends on genes and proteins (Hoffmann and Valencia). The importance of author ambiguity has also been addressed before by the WikiAuthors initiative.

Results and Discussion
Geographical Distribution
Here are several aspects which are not very clear from the analysis carried out by the authors: how the authors group / define geographical regions and whether the location of the annual meetings has a significant effect on the number of publications from certain (geographically close to the meeting venue) regions. When the authors mention the top five countries, they should keep in mind that EU-15 doesn't correspond to a country. It is also not clear why they did not dissect Europe into the corresponding countries in this study.

Basic statistics and demographics
This section lack a clear focus, many aspects are mentioned but it is not clear what the main idea the authors want to present in this section would be. The relevance of some of the mentions points is also not clear.

Analysis of the co-authorship graphs
The authors did a detailed study of the main characteristics of the resulting co-author network with interesting results. What is missing here is a clear motivation why this analysis is relevant , something which is missing in the current manuscript.

Topic modelling
Although the presented approach is interesting and LSI has been used successfully in terms of biomedical literature processing in the domain of microarray data analysis some aspects are not clear form the paper. It seems that the authors used stemmed words as basic input for their system. It seems that using only single words and not exploring additional aspects such as collocations, technical terms etc,.. can only show a partial picture of the relevant topics mentioned in the used abstract collection.

Topic clusters
This is a specially interesting part of the presented work and would benefit from more details.