On 2016 Sep 27, UFRJ Neurobiology and Reproducibility Journal Club commented:

We’d like to point out that in the legends of all figures in these article (except for Figure 7), the authors mention that data is presented as mean +- s.e.m.; nevertheless, in some of these figures (e.g. 2F, 3B, 3C and 3F), the error bars are clearly asymmetric. As standard errors of the mean are symmetric by definition, this seems to indicate that the data does not match the description in the legend. Therefore, the authors should clarify what the error bars presented stand for, as different kinds of error bars represent very different things and can lead to different inferences about the data (Cumming G, 2007; Krzywinski M, 2013).

On 2016 Sep 27, UFRJ Neurobiology and Reproducibility Journal Club commented:

We’d like to point out that in the legends of all figures in these article (except for Figure 7), the authors mention that data is presented as mean +- s.e.m.; nevertheless, in some of these figures (e.g. 2F, 3B, 3C and 3F), the error bars are clearly asymmetric. As standard errors of the mean are symmetric by definition, this seems to indicate that the data does not match the description in the legend. Therefore, the authors should clarify what the error bars presented stand for, as different kinds of error bars represent very different things and can lead to different inferences about the data (Cumming G, 2007; Krzywinski M, 2013).