Like many of you, I've submitted countless manuscripts and grant applications that have received mind-bogglingly absurd suggestions/corrections/critiques from so-called experts that were assigned to peer-review the work. Shockingly, it somehow didn't matter to the editors or program officers that many of these brilliant critiques were already contained within the article/grant, were just flat-out wrong, or asked for additional experiments that would likely produce several Nobel Prizes. Thus, my frustrations with peer review has led me here--to offer an outlet for response in an unstructured, humerous way.
This is the deal. Authors are encouraged to submit their responses to peer-reviewed manuscripts and grant applications. And I don't mean the politically correct, carefully crafted scientific responses that you send to journal editors or granting agencies. Responses should take no prisoners. Sarcasm is encouraged. Foul language too if it assists in making the point. Anything that helps to relieve the frustration and release the tension. I only ask that you avoid racist, sexist, or homophobic remarks and remove any potential personally identifying information. All else is fair game. To kick off this endeavor I've posted my "response letter" from a recently rejected article. Enjoy!
This article was peer-reviewed by 3 "experts". I removed a lot of the bloviating by Reviewers 1 and 3 who seemed intent on making themselves feel smart by going on for paragraphs about how much they know about the topic of interest. My responses to the highlighted text from the review are in bold red text :
Reviewer #1:
Overall, this is a very interesting article with some intriguing results. However, I do not feel that the authors have a completely convincing case to back up their bold claims of dissociation because of their rather limited choice of behavioral tests and interpretation.
1. The use of a mouse with complete 5-HT depletion is a very useful start; although it would be of more interest ultimately to determine the functions of 5HT in discrete neural systems by use of conditional knock-downs. The authors should perhaps point out this short-coming. However, I do concede the great usefulness of this preparation in assessing the neurochemical basis of the effects of SSRIs. You’re so kind, jerk-off.
2. In general, the authors have employed easy-to-implement observational tests, which are useful and suggestive, but can be limited in interpretation. Tell that to the several thousand researchers throughout the world that use these “easy to implement observational tests”. Come to think of it, you should alert every publisher of scientific articles about this too. It’s an epidemic! You need to stop everyone from using these tests!
3. Overall, the authors' definitions of impulsivity and compulsivity are vague and not very well formulated. This aspect of the Introduction needs improvement, as some tests are simply asserted to be tests of impulsivity or compulsivity, seemingly on the basis of anthropomorphic subjective impression. The Chairman of Psychiatry wrote these definitions but I guess he’s not knowledgeable enough to do so? Please enlighten us all-knowing, anonymous piece of shit that you are.
4. This apparent lack of theoretical sophistication leads the authors into some difficult paradoxes. For example, novelty suppressed eating is almost universally agreed to be a test of anxiety (conflict between approaching/eating and freezing), and yet these authors call it a test of impulsivity. Similarly for the light/dark tests. The dark emergence test is what we used, chief. In fact, the rationale of these tests is not so different from that of the elevated plus maze. The authors conclude from their pattern of findings that 5-HT is implicated in impulsivity but not anxiety, as they label the elevated Y maze???? test as a test of anxiety and the other two as tests of impulsivity. Not True considering we did not use the Y maze. You are embarrassing yourself here.
1. The use of a mouse with complete 5-HT depletion is a very useful start; although it would be of more interest ultimately to determine the functions of 5HT in discrete neural systems by use of conditional knock-downs. The authors should perhaps point out this short-coming. However, I do concede the great usefulness of this preparation in assessing the neurochemical basis of the effects of SSRIs. You’re so kind, jerk-off.
2. In general, the authors have employed easy-to-implement observational tests, which are useful and suggestive, but can be limited in interpretation. Tell that to the several thousand researchers throughout the world that use these “easy to implement observational tests”. Come to think of it, you should alert every publisher of scientific articles about this too. It’s an epidemic! You need to stop everyone from using these tests!
3. Overall, the authors' definitions of impulsivity and compulsivity are vague and not very well formulated. This aspect of the Introduction needs improvement, as some tests are simply asserted to be tests of impulsivity or compulsivity, seemingly on the basis of anthropomorphic subjective impression. The Chairman of Psychiatry wrote these definitions but I guess he’s not knowledgeable enough to do so? Please enlighten us all-knowing, anonymous piece of shit that you are.
4. This apparent lack of theoretical sophistication leads the authors into some difficult paradoxes. For example, novelty suppressed eating is almost universally agreed to be a test of anxiety (conflict between approaching/eating and freezing), and yet these authors call it a test of impulsivity. Similarly for the light/dark tests. The dark emergence test is what we used, chief. In fact, the rationale of these tests is not so different from that of the elevated plus maze. The authors conclude from their pattern of findings that 5-HT is implicated in impulsivity but not anxiety, as they label the elevated Y maze???? test as a test of anxiety and the other two as tests of impulsivity. Not True considering we did not use the Y maze. You are embarrassing yourself here.
However, even then, one has to explain the strange pattern of effects in these three tests. What is strange? The mice are not anxious but are impulsive. In fact, in the light-dark test they spend more time in the light. It all fits in my mind. But clearly I’m delusional thinking that these results fit a common phenotype and that it could get published.
5. To call the forced swim test and the tail suspension test assays for depression is also dubious. This calls into question perhaps tens of thousands of papers on the topic of depression in rodents, many in more high profile journals than this one. What these tests do well is predict effects of anti-depressant drugs - they do not necessarily index depression, a distinct from some (rather ill-defined) measures of stress. Even the paradigm from which they are derived (learned helplessness) and which has a far sounder theoretical basis, is no longer regarded simply as a test of depression. I agree it was worth running these tests and the lack of effects of 5-HT depletion is interesting and probably important- but this may say more about how SSRIs work than how 5-HT contributes to depression. I would also liked to have seen effects on tests based on other aspects of depression, such as anhedonia. Did you even read the fucking paper, asshole? You are an absolute disgrace.
6. The effects of fluoxetine are among the most important findings in the study. However, it would have been useful to know if the 'paradoxical' therapeutic effects of the drug were nevertheless not blocked by 5-HT receptor antagonists, in order to enhance confidence in the non-5-HT interpretation of mechanism. Exploring the non-5HT mechanism would also have been important, although that should probably more properly be a topic for another study. At least you cut us some slack there, Richard.
7. raphe should be spelt raphé. Thank you Mr. Merriam-Webster.
Reviewer #2: This is a very sound paper. However, there are some issues that need to be addressed.
1. There is evidence that compulsive behaviors might be related to striatal dysfunctions. The authors should provide data on the effects of tryptophan on serotonin and 5-HIAA on the striatum. See the tryptophan load studies in the paper we referenced, lazy ass. In addition, they should provide information about the impact of serotonin depletion on other monoaminergic systems that are known to interact with serotonin.
2. The authors failed to explain how fluoxetine was able to improve behaviors in these animals. They should show if fluoxetine might impact post-synaptic 5-HT system even in the absence of 5-HT. This is clearly not the focus of this paper. Even Richard #1 agreed on that. In addition, many of these drugs are known to influence BDNF levels. The authors should provide details on the effects of fluoxetine on BDNF and/or other neurotrophins. Recently pathological changes in BDNF have been reported in several neuropsychiatric disorders, it is therefore important for the authors to provide more details on BDNF expression in the knockout mice, especially since they are responding to a class of drugs that have been shown to impact BDNF levels. Ok, we get it. You study BDNF in neuropsychiatric disease. Calm down. We’ll do it.
3. The discussion is too long. It needs to be more concise. Really? Such a helpful comment. But unfortunately your review is too concise and we don’t know where we should cut it down without any real feedback. Thanks for taking 5 minutes (maybe?) to review this. You better get back to those BDNF studies quickly (I’m sure you already did).
4. I am not sure that the colors add much to the paper. That’s a shame b/c we wanted to make the text pink in honor of breast cancer awareness month. I guess that’s out of the question.
Reviewer #3
5. To call the forced swim test and the tail suspension test assays for depression is also dubious. This calls into question perhaps tens of thousands of papers on the topic of depression in rodents, many in more high profile journals than this one. What these tests do well is predict effects of anti-depressant drugs - they do not necessarily index depression, a distinct from some (rather ill-defined) measures of stress. Even the paradigm from which they are derived (learned helplessness) and which has a far sounder theoretical basis, is no longer regarded simply as a test of depression. I agree it was worth running these tests and the lack of effects of 5-HT depletion is interesting and probably important- but this may say more about how SSRIs work than how 5-HT contributes to depression. I would also liked to have seen effects on tests based on other aspects of depression, such as anhedonia. Did you even read the fucking paper, asshole? You are an absolute disgrace.
6. The effects of fluoxetine are among the most important findings in the study. However, it would have been useful to know if the 'paradoxical' therapeutic effects of the drug were nevertheless not blocked by 5-HT receptor antagonists, in order to enhance confidence in the non-5-HT interpretation of mechanism. Exploring the non-5HT mechanism would also have been important, although that should probably more properly be a topic for another study. At least you cut us some slack there, Richard.
7. raphe should be spelt raphé. Thank you Mr. Merriam-Webster.
Reviewer #2: This is a very sound paper. However, there are some issues that need to be addressed.
1. There is evidence that compulsive behaviors might be related to striatal dysfunctions. The authors should provide data on the effects of tryptophan on serotonin and 5-HIAA on the striatum. See the tryptophan load studies in the paper we referenced, lazy ass. In addition, they should provide information about the impact of serotonin depletion on other monoaminergic systems that are known to interact with serotonin.
2. The authors failed to explain how fluoxetine was able to improve behaviors in these animals. They should show if fluoxetine might impact post-synaptic 5-HT system even in the absence of 5-HT. This is clearly not the focus of this paper. Even Richard #1 agreed on that. In addition, many of these drugs are known to influence BDNF levels. The authors should provide details on the effects of fluoxetine on BDNF and/or other neurotrophins. Recently pathological changes in BDNF have been reported in several neuropsychiatric disorders, it is therefore important for the authors to provide more details on BDNF expression in the knockout mice, especially since they are responding to a class of drugs that have been shown to impact BDNF levels. Ok, we get it. You study BDNF in neuropsychiatric disease. Calm down. We’ll do it.
3. The discussion is too long. It needs to be more concise. Really? Such a helpful comment. But unfortunately your review is too concise and we don’t know where we should cut it down without any real feedback. Thanks for taking 5 minutes (maybe?) to review this. You better get back to those BDNF studies quickly (I’m sure you already did).
4. I am not sure that the colors add much to the paper. That’s a shame b/c we wanted to make the text pink in honor of breast cancer awareness month. I guess that’s out of the question.
Reviewer #3
1. The genetic background of the mice is never stated which is not acceptable. It is not sufficient to cite this by reference to another paper. That’s news to us. Maybe you should also tell this to the several thousand researchers, editors, and journals that have made this practice commonplace. Or perhaps we missed your memo about this? Sorry about that. The authors must state the original genetic background of the mice in which homologous recombination was done and if a mixed background, such as 129Sv x C57BL or similar, it must be stated how the current mice were derived. Where the mice in these experiments backcrossed from the founders to one of these strains? If so which strain (and substrain) was it? And if backcrossing was done, how many generations were done to reach the ones in the present experiments? Are all the mice in the experiments in the paper from the exact same backcross generation or from multiple generations? Thank you, Gregor Mendel. Your knowledge of genetics is superior to ours. You win.
2. It is well established that it is problematic to do null x null and WT x WT breeding and compare offspring of the two, but we are told nothing of the breeding strategy used in this work. That’s right, you had to look up the reference. Why re-invent the wheel? And where did you come up with the KO x KO and WT x WT thing? Did you pull that from the last grant you trashed err… reviewed of ours? Funny because that should not come into play here at all. Oh and by the way, Jackson Labs regularly employs this strategy, but hey, what do they know? It is generally accepted practice that the optimal approach is to do het x het breeding and select null and WT offspring from each litter and discard or use otherwise the het offspring (as for colony maintenance, etc.). Please specify what breeding strategy was used to generate the mice used in these experiments. Again, interesting that you would concern yourself so much with this. Do you have some previous knowledge of our mouse breeding strategy that you don’t like? I guess you’re too stupid to realize that we know exactly who you are from your comments given that they are identical to the grant you reviewed.
5. Statistical presentation: Add F-ratios, df's, and p-values. If the authors do not want these in the text, then put them in the figure captions. Here’s one F of value that I forgot to add: Go F yourself.
11. For example, they dismiss evidence that 5-HT plays a neurotrophic role during brain development despite the fact that the evidence supporting this is extensive while the evidence against it is limited coming almost exclusively from mouse consitutive KO approaches (Trp2, Pet-1, SERT disruption) . These models have limitations and are far from definitive. Yes, but Nature seems to love them and has no problem promoting them as definitive in their journals. I guess you’ll have to tell all of the authors of those papers that they are wrong about everything they’ve published in the past few years. It is entirely plausible that the absence of 5-HT during early brain development can be compensated for such that the brain reorganizes and can function relatively normally but when 5-HT is present early it becomes integrated into normal brain function such that interruptions of established signaling causes dysfunction (as in depression and anxiety disorder). So you know exactly how disruptions in 5HT signaling contribute to depression and anxiety. I guess we should all just quit and go home now. You’ve solved it. When are you going to let everyone else know?
This is not to suggest that evidence that antidepressants have other, non-neurotransmitter reuptake mechanisms (such as changes to BDNF, neurogenesis, or direct receptor effects) are not contributing to how these drug work, but rather that the current evidence in Tph2 constitutive KO mice is insufficient to reach a broad conclusion that 5-HT does not have a developmental role or that it is not important in mature brain in relation to the onset of anxiety or depression given that these disorders are episodic and may involve fluctuations in 5-HT signaling over time. Take a breath, please. This is the longest sentence I’ve ever read outside of Faulkner. And yet I still don’t know what the fuck you are trying to say. Then again, I read The Sound and the Fury 6 times before I even knew what The Sound was. It seems possible given the complexity of the data now available on 5-HT in brain and its role during brain development that we are only now able to see how complex and subtle its effect really are. The authors' data do not permit them to dismiss alternate explanations so blithely and they should revise their Discussion accordingly. Did we dismiss them blithely? Oh, I’m sorry we meant to dismiss them abso-fucking-lutely because they all suck and are not supported by the facts. Sorry for the confusion.
12. P-18, Par-2, L-3: "To best of our knowledge" should be "To the best of our knowledge". “To the best our knowledge”, you should not be reviewing this paper because of clear conflicts of interest that you refuse to acknowledge. But I guess it’s easier to destroy your competitor first-hand than to be honest with the editor.
12. P-18, Par-2, L-3: "To best of our knowledge" should be "To the best of our knowledge". “To the best our knowledge”, you should not be reviewing this paper because of clear conflicts of interest that you refuse to acknowledge. But I guess it’s easier to destroy your competitor first-hand than to be honest with the editor.
