Mitra and Vyas

rmav.jpg

Rupshi Mitra and Ajai Vyas are a husband-and-wife team at Nanyang Technological University.

Screen Shot 2021-03-22 at 8.49.23 AM.png

I analyzed research and academic supervision work by Mitra and Vyas since they came to NTU, please see Part II of the report. For a decade or so, here is what Mitra and Vyas have been doing at NTU:

Mitra's science

  • As NTU faculty, Mitra has been addressing one research project: the effect of environmental enrichment on stress in rats. Environmental enrichment means putting the rats in big cages with swings and toys. It became fashionable some time ago, the idea behind it is that as animals with complex social brains, rats and humans would cope better with stress if they are in a physically, mentally, and socially stimulating and varied environment. There are criticisms, of course, not least of which nutritious food available to the enriched rats is not available to the control rats, so we do not know how much of any effect we measure is because of environmental enrichment or because of something they are eating (nutrition). You can have the opinion, as others do, that the hype around environmental enrichment is not all what it is made to be - in short, it will not solve our social problems with regards to stress.

  • A stress paradigm Mitra uses is brutal: 2 hours of chronic restraint every day for ten days. This means the rat is put in a cylinder where it has no room to move. Studies have shown that just 10 minutes or even less is enough to seriously and chronically stress a rat.

  • Mitra starts the stress paradigm and environmental enrichment on the same day and for two weeks. This means that we cannot tell if any effect on stress we are seeing is due to environmental enrichment. It's a different question. If we were looking at the effect of enrichment on stress, then obviously we would start the enrichment a period of time, say a month or two, before we start stressing the rats.

  • When Mitra looks at what happens in the brain cells, she uses a drawing tube to draw the neurons. This means that someone sits down at the microscope and uses an outdated device to hand-draw the neurons, one by one. Mitra published articles from NTU in which a drawing tube was used for analysis of neuronal structure up to 2020, not half a century ago. Note the articles refer to a camera lucida, which is a drawing tube. Obviously, we should be using a digital camera to take a photo of the neurons, which we would then put into a powerful software, available at NTU, to analyze neuronal structure. Incidentally, I never actually saw the famous drawing tube at Mitra's lab in NTU, nor any x100 objective microscope lens she reports using for analysis of dendritic spines (counted manually of course).

  • When Mitra looks at what happens to rat behavior, she forces the researchers to use paper-and-pencil and a stopwatch. Needless to say, there are a number of computer programs available to instantly analyze the behavioral experiments.

  • Mitra reports doing experiments in animals that are neither alive nor dead, putting data from one experiment into another, using antibodies that do not and never did exist, and other bizarre things, see the report for more.

Vyas' science

  • As faculty at NTU, Vyas has been looking at the effect of Toxoplasma gondii on fear behavior in rodents. This is the parasite that goes from cats to humans, rats, and mice, and back. Apparently, it can alter rat (and mouse?) behavior to make them 'fearless' of cat smell (normally, rats and mice are born with a healthy fear, terror even, of anything that smells like a cat, or a predator which will eat it). There is hype around this topic, also criticized, see the reviews by Doherty and by Worth and others.

  • Vyas' hypothesis is that T. gondii, the parasite, increases testosterone production in rats. This increased testosterone then induces epigenetic changes in an infected rat's brain, which makes it fearless of cats and their smell.

  • We are not really sure how many rats infected with the parasite actually go on to become fearless. From Vyas' experiments it appears all infected rats become fearless. It's probably much, much, much less than that. This is important because if we don't know and are not unambiguously told what percentage of rats infected with the parasite become fearless, the results are called into question.

  • We are not really sure if mice react to the parasite the same way. Vyas told us in a review that the effect of the parasite is highly specific to rats - it will never work in mice. But then later he recreated the effect in mice, only without the parasite. In other words, Vyas tells us that the mechanism by which the parasite makes rats fearless is the same in rats and mice, and at the same time tell us that the parasite-testosterone-fearless effect only applies to rats, never to mice.

Vyas' and Mitra's statistics

In their publications and when it comes to statistics, Vyas and Mitra engage in the following (see Problematic ethics for more):

  • Remove 'outliers' and don't tell us they did (outliers are in theory data that are far from average and we can't explain, in practice data points that skew the result against our hypothesis).

  • Invariably use a kind of statistics where you think you know where the result is going to before you do the experiment. That's a bit unusual in this kind of work for several reasons. Usually in animal physiology kind of thing, we just use the sort of statistics where we do not really make assumptions about where the result is going. I guess this allows Vyas and Mitra to play around with 'significance', the statistical way of saying 'the data proves the hypothesis'.

  • Mix 'effect size' and 'significance' at will. 'Effect size' is another statistical parameter distinct from 'significance' and is used to describe the 'bigness' of difference between different experimental results. The problem is that Vyas and Mitra refer to effect size when it is in line with the hypothesis and ignore it when it isn't, and the same for significance.

  • Use a 'two-by-two' design to duplicate data in publications. In one experiment you might have two 'controlled' variables and two variables you are measuring. If you chart it, you get a big square divided into four smaller ones. That's one experiment, one study. What Vyas and Mitra do is put one little square in one publication, another little square in another publication, and so on.

Vyas' and Mitra's coauthors and students

When looking at their publications, it is difficult or impossible to tell who did what, where, and when. For example, a doctoral student of Vyas' may say he did all this work, but then the article is published by someone else, someone who might not even have been in Singapore when the work was allegedly done. A lot of work in the theses supervised by Vyas and Mitra is repeated, just the same dubious material over and over again. See the report.

Peer review for Vyas and Mitra

In addition to duplication, misrepresentation, falsification, and fabrication, some articles by Vyas and/or Mitra appear to have been edited or reviewed under suspicious circumstances. For example, an editor for an article by Mitra in Frontiers had the article 'reviewed' by two researchers who share/shared affiliation with him and are co-authors of his as well as each other on more than two dozen articles. This is not OK, you do not get your close collaborators and co-authors, or 'friends', and only your friends, to review an article, especially when the publisher guidelines explicitly say you may not.

Final Year Projects supervised by Vyas and Mitra

A Final Year Project at NTU is like a bachelor degree project or thesis. Many students supervised by Mitra and Vyas were clearly demoralized. One very clever and competent student of Vyas' got what appears to be solid results trashing Vyas' hypothesis on increased testosterone and what-not. He was not allowed to show these results. A student of Mitra's got what appears to be incredibly robust data which goes against Mitra's ideas on the effect of environmental enrichment on brain plasticity. This student's data was deliberately ruined and she was not allowed to present the original data. See the report for more.

Illegal animal experiments

Vyas and Mitra kill animals without anesthesia for no reason, use the same data between experiments and publications, misreport how experiments were done, and very rarely (almost never) report Animal Use Protocol numbers. In one interesting set of experiments, a rat gene was inserted into mouse to alter what should have been the mouse gene. I wrote to the publisher of this article. The publisher of this article told me to ask Vyas about that, and apparently closed the case. In correspondence with the publisher Vyas did not have an answer of course, unless it was something he wrote about my 'opinion'. In other words, Vyas believes that using a rat gene in mouse to alter the mouse gene in a functional manner is a matter of 'opinion' as opposed to scientific fact.