But are there depression jeans?
May 19, 2019 5:54 AM   Subscribe

There is no "depression gene". The authors go on to demolish every other "depression gene" connection in the existing literature. They went after the lot. Nothing. No clear evidence for any given gene, in any polymorphic form, with any effect on depression, as either measured by itself or in combination with any other environmental effect. Paper (paywalled).

Slate Star Codex has a deeper dive into the issues raised by the paper, which got an endorsement from one of the paper's authors:
They explain that given what we now know about polygenicity, the highest-effect-size depression genes require samples of about 34,000 people to detect, and so any study with fewer than 34,000 people that says anything about specific genes is almost definitely a false positive; they go on to show that the median sample size for previous studies in this area was 345.

They show off the power of their methodology by demonstrating that negative life events cause depression at p = 0.000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001, because it’s pretty easy to get a low p-value in a sample of 600,000 people if an effect is real. In contrast, the gene-interaction effect of 5-HTTLPR has a p-value of .919, and the main effect from the gene itself doesn’t even consistently point in the right direction.
(One of the responses to the endorsement has a useful caution about Slate Star Codex: "Scott Alexander is a psychiatrist with good posts on psychiatry, okay posts on economics, bad posts on politics and terrible posts on AI.")

Sketchy sci-hub link to paper.

The same lab got similar results with schizophrenia candidate genes: No Evidence That Schizophrenia Candidate Genes Are More Associated With Schizophrenia Than Noncandidate Genes.

What we've learned over the past few years about the connection between specific genetic variations and behaviour is summed up in The Satan Gene.
posted by clawsoon (65 comments total) 36 users marked this as a favorite
 
Statistically not surprising. Reliability of scientific publishing (not even talking about scientific reporting in non-scientific press, which is abysmal) got a lot of press a couple of years ago and as far as I know not much has improved. There are some scientists working on doing science better (retraction watches, hypotheses escrow/registries, and more funding and publishing for reproducibility research) but I wouldn't claim these initiatives have reached the critical mass they need to.

Many scientists are firmly bitten by confirmation bias and don't even have the training to recognize the problem, let alone fix it.
posted by kalessin at 6:05 AM on May 19 [6 favorites]


So not so much "there is no depression gene" as "no study done so far could have validly identified one."

I’m glad to see someone out there making the point. Way too many pop-sci articles get based off half-baked research.
posted by Tell Me No Lies at 6:40 AM on May 19 [22 favorites]


So not so much "there is no depression gene" as "no study done so far could have validly identified one."

There are almost certainly /lots/ of genes involved in depression that each individually have an extremely weak effect.
posted by pharm at 7:24 AM on May 19 [11 favorites]


Isn’t Slate Star Codex some kind of eugenicist, “racial rationalist” or similar Intellectual Dank Web type?
posted by acb at 7:55 AM on May 19 [15 favorites]


There are almost certainly /lots/ of genes involved in depression that each individually have an extremely weak effect.

Agreed. But there’s been no research done yet that would confirm or deny that.
posted by Tell Me No Lies at 8:05 AM on May 19 [1 favorite]


acb: Isn’t Slate Star Codex some kind of eugenicist, “racial rationalist” or similar Intellectual Dank Web type?

It (previously) is definitely problematic.
posted by clawsoon at 8:08 AM on May 19 [7 favorites]



not even talking about scientific reporting in non-scientific press, which is abysmal

The Science News Cycle
posted by lalochezia at 8:11 AM on May 19 [2 favorites]


I wonder if there's going to be some sort of backlash against psychiatric medicines at some point. Look at how much traction anti vaxxers have got protesting against things which provably work. The only consistent effects of psych meds are the side effects. They're a big sitting target.
posted by thatwhichfalls at 8:14 AM on May 19 [3 favorites]


Is the converse side of this the idea that no genetic variations can protect you from depression? Everybody hurts, sometimes?
posted by clawsoon at 8:20 AM on May 19 [2 favorites]


The only consistent effects of psych meds are the side effects.

Well and the whole stopping people from killing themselves, alleviation of symptoms, ability to get on with your day without constantly sobbing, lessening or removing of panic attacks... I could go on.

The problem with psych meds is that yes, quite a few do have side effects that range from minor to major- and that a lot of prescribing docs don't give enough time and attention to their patients so they can fiddle around and find the right one. They're a very individual type of medicine- what works for me won't necessarily work for you, and vice versa. I'd argue we have a doctor problem rather then a med problem. Not to mention- there already *are* people protesting psych meds, the Scientologists.
posted by Homo neanderthalensis at 8:29 AM on May 19 [61 favorites]


And just because there is no *one* gene that predicts depression (that we know of so far) doesn't mean it doesn't run in families. Whether that's because of a constellation of genes or generational trauma of course is up for debate.
posted by Homo neanderthalensis at 8:30 AM on May 19 [18 favorites]


That's a very broad brush, thatwhichfalls. Where are you getting your information?
posted by Selena777 at 8:42 AM on May 19 [4 favorites]


Isn’t Slate Star Codex some kind of eugenicist, “racial rationalist” or similar Intellectual Dank Web type?

Yeah basically, but if he sticks to his field and stays specific he's probably okay.

(Unless he's advocating for medicines that are more efficacious but also have higher mortality rates, where the skull face starts to show a little bit?)

In this article I like where he talks about the social mechanisms by which this gigantic failure of science has happened -- yet at another time, if you said western science has western culture in it, he would give himself hiccoughs laughing at you derisively because people don't begin their research papers w/ invocations of Thor.
posted by fleacircus at 8:48 AM on May 19 [2 favorites]


I mean. Psych meds saved my life. In fact, scary highly-stigmatized psych meds that are often given in abusive ways to at-risk populations saved my life. I am not in any way antipsychiatry, even for meds and conditions that it would be pretty easy to be antipsychiatry about.

But. Most psych meds do quite poorly in clinical trials. Let's take that fact at face value. It isn't inexplicable. It doesn't mean they're useless. There are some great alternative hypotheses about what's causing it, including stuff like "Choosing the right psych med for a specific patient is hard." But it is a fact — psych meds struggle to get significant results and tend to show very small effect sizes.

Given that, yeah, I'm also surprised there isn't more backlash than there is.
posted by nebulawindphone at 8:50 AM on May 19 [28 favorites]


Does the neutral theory apply to protein sequences, not just DNA sequences? Maybe our genes have evolved not just to do their jobs, but to be robust to mutations, so that - with a few dramatic exceptions - if you swap out one amino acid for another, you still get a protein that works reasonably well? Selection for protein robustness?
posted by clawsoon at 8:51 AM on May 19


I've treated a disturbingly large number of people destabilized because they paid out of pocket for expensive gene tests then had their medications changed or stopped. The results were not good. I'd be hard pressed to think of a single example where a gene test made a positive difference (specific liver metabolism tests are more useful). Gene tests in psychiatry are (currently) 99% menace. The companies marketing these to consumer are basically Theranos-lite scammers.
posted by meehawl at 8:53 AM on May 19 [21 favorites]


I've heard about mixed results from studies with antidepressants, but that's only one category of psych med, and there's categories within that category. How do powerful yet ineffective drugs get through clinical trials?
posted by Selena777 at 8:53 AM on May 19


ADs prescribed to people with properly diagnosed MDD do indeed work (sadly not for me, but I'm glad for the others).
Most ADs are prescribed to people who do not have MDD, in which case they are little more than evil sugar pills.
Yes it's a doctor problem and can be fixed but that's going to take a generation at least.
I'm not advocating for a backlash. I know people who've been helped by these drugs. I'm simply observing that treating them as happy pills is not sustainable. People will eventually notice.
posted by thatwhichfalls at 8:55 AM on May 19 [2 favorites]


No gene for depression? Just as we thought, it's a moral failing. sarcasm, of course, but in these times, might as well label it.
posted by theora55 at 9:04 AM on May 19 [10 favorites]


Most psych meds do quite poorly in clinical trials. Let's take that fact at face value. It isn't inexplicable. It doesn't mean they're useless.

Are you claiming that about the widely produced and used ones, or about the experimental ones? Because, yes, clinical trials of proposed new anti-depressants turn out neutral all the time. So do clinical trials of proposed cancer drugs and a bunch of other things. Making new medication is hard.

The well-distributed anti-depressants however showed distinct improvements in mood. In several cases researchers weren't even looking for anti-depressants but that quality showed up as a side effect of existing medication. As for the others -- well, you could claim the people running the trials were firmly in the pocket of big pharma, but if that's the case why are there so many failed trials?

Last but not least there are millions of test cases for the well-distributed anti-depressants. To claim they don't work is to claim that millions of people -- many of whom are working with therapists that are professionally adept at judging mood -- are fooling themselves, which is a very bold thing to do. So shine on crazy diamonds, but realize that you've got quite a task in front of you if this a hill you want to die on.
posted by Tell Me No Lies at 9:07 AM on May 19 [3 favorites]


Dude. I'm specifically not claiming they don't work, and I made that abundantly clear. I'm claiming they have a PR problem.
posted by nebulawindphone at 9:10 AM on May 19 [5 favorites]


Tell Me No Lies: To claim they don't work is to claim that millions of people -- many of whom are working with therapists that are professionally adept at judging mood -- are fooling themselves, which is a very bold thing to do. So shine on crazy diamonds, but realize that you've got quite a task in front of you if this a hill you want to die on.

I thought that there were already some meta-analyses which have done the hard work. I seem to recall that a minimum of 80% of the effect of both therapy and depression medications is a result of the placebo effect, and that might be an underestimate because of piercing-the-veil effects. Is my info on that out-of-date?
posted by clawsoon at 9:16 AM on May 19 [3 favorites]


Well, I'm just going to go ahead and say I knew all along that it was because I have to live in this society.
posted by Vulgar Euphemism at 9:21 AM on May 19 [18 favorites]


I also thought it was pretty well established that anti-depressants have effectiveness above placebo only for moderate/major depression? Also that the seratonin hypothesis is questionable. So basically all they know is that the meds work somewhat for the more severe cases, but not why.
posted by schwinggg! at 9:21 AM on May 19 [3 favorites]


Is my info on that out-of-date?

Yes. What finally got me to stop lurking and make an account here was to talk about the seminar I'd recently attended where the old "antidepressants mostly don't work except for severe cases" was revealed to be a statistical error that got picked up and widely, widely, widely reported because it fed into the rampant med-skep mentality our society has. (Ok that last part is me editorializing.)

But no, the antidepressants studied has highly significant positive effects on depression of all severities.
posted by traveler_ at 9:24 AM on May 19 [31 favorites]


What finally got me to stop lurking and make an account here was to talk about the seminar I'd recently attended where the old "antidepressants mostly don't work except for severe cases" was revealed to be a statistical error that got picked up and widely, widely, widely reported because it fed into the rampant med-skep mentality our society has.

Do you have a link to anything about this? Pretty much everything I've read on the subject seems to support the idea that antidepressants don't do well in clinical trials and especially not for mild depression. I'd be interested to see anything to the contrary.
posted by armadillo1224 at 9:35 AM on May 19 [3 favorites]


This appears to be the biggest recent meta-analysis of depression medications. Here and here are articles about it.
posted by clawsoon at 9:41 AM on May 19 [15 favorites]


Does the neutral theory apply to protein sequences, not just DNA sequences? Maybe our genes have evolved not just to do their jobs, but to be robust to mutations, so that - with a few dramatic exceptions - if you swap out one amino acid for another, you still get a protein that works reasonably well?

The term for this kind of built-in robustness is degeneracy, and while it's most commonly applied in the context of codons, it's also sometimes used to describe other levels in biological systems:
The genetic code relating sequences of polypeptides and polynucleotides is degenerate because there are many more triplet codons than encoded amino acid residues. Consequently, an enormous number of structurally distinct mRNA species could be translated to generate the amino acid sequence of any particular protein. This degree of structural variation can be considered to be merely the tip of the iceberg, particularly if we broaden our definition of degeneracy to include variations in polynucleotide sequences that result in functionally equivalent gene products. For example, it is now clear that at many sites along the polypeptide chain, substitution of one amino acid residue for another has little effect on overall protein conformation or function. By inference, we can assume that an astronomical number of different amino acid sequences could contribute equally to the survival of the species.
Which is not to say that all amino acid substitutions are equal. Proteins need to fold at specific places and have certain catalytic sites in order to be functional. So while many substitutions have negligible effects, others can be quite significant.

How do powerful yet ineffective drugs get through clinical trials?

Newly approved drugs generally have to be equally effective or better than existing treatments in order to get approved, or else fill some new niche. For some conditions, the efficacy of existing treatments is low, so the threshold for new drugs is also low. You see this a lot in oncology, where newly approved treatments may only extend life by a few weeks or months on average. Antidepressants, regardless of mechanism, tend to have fairly low rates of efficacy compared to other classes of drugs, and within-class they are all roughly comparable on a population level.
posted by dephlogisticated at 9:47 AM on May 19 [7 favorites]


Do you have a link to anything about this?

I don't sorry. I haven't been able to find my notes from that seminar, or google up the papers he published on it. And the topic of his seminar wasn't really on that specifically, but on the general problem of statistical mistakes common among scientists. That paper was given as an example of statistics used in meta-analyses that don't measure what people think they measure, and in this case a form of Simpson's paradox resulted.
posted by traveler_ at 10:16 AM on May 19 [1 favorite]


As a non-expert but someone who has been navigating mental healthcare as a patient for two decades now: I think there is also an issue about What We Talk About When We Talk About Depression.

A lot of studies on drug efficacy are based on major depressive disorder. However, most outpatients are actually being treated for atypical depression, not MDD.

On a practical level, that means a whole bunch of people with atypical depression are medicated based on what seems to work for MDD, even though the two respond differently to medication.

Again, definitely not an expert! If the above is incorrect or you can point me to recent studies regarding the difference in drug efficacy between the two diagnoses please let me know.
posted by evidenceofabsence at 10:47 AM on May 19 [4 favorites]



I wonder if there's going to be some sort of backlash against psychiatric medicines at some point.


another one? there's been so many, both from inside and from outside the mental health professions. led by everyone from doctors at the extremely respectable end to scientologists at the other end. even among those who shout down criticism of existing pharmaceutical treatments for mental illness, it is still much, much more socially acceptable to refuse psych meds for personal reasons than to refuse vaccines, which is as it should be. what more do you want from a backlash?
posted by queenofbithynia at 10:50 AM on May 19 [16 favorites]


Reading Sapolsky's Behave and he says that in general the provable relatedness (trying to avoid statistic-speak because I are not a statistician despite getting an A in it and having a doctorate) for individual genes related to mental illness is very low even in studies where it is shown to have an effect, but some genes in interaction with environmental variables are much more likely to be related and many genes may go into having any effect on mental illness.
posted by Peach at 11:26 AM on May 19 [4 favorites]


I can't find it now, but I was reading someone recently who pointed out that in the case of depression, "placebo" isn't just a sugar pill, it's a sugar pill within the context of an intervention by a whole bunch of people who are trying to help. For some cases of depression, it seems like knowing that people are trying to help you could, by itself, make a difference, though I don't know the science on it.
posted by clawsoon at 11:27 AM on May 19 [7 favorites]


No more backlash, please, as a severely ill person living in the world*, it's there. It's bad! It is here, on this website, in this very thread! It's not always noticeable because that is how ableism operates! If the full force of it hasn't hit you (and you're not taking drugs that even advocates viciously stigmatize) be grateful.

And ugh, SSC, I would take literally anything Scott Alexander says, and yes up to and including the psychiatry stuff, with a huge like, pillar of salt. I would also consider how often we talk about side effects and jump on studies and drugs not working and etc in a very specific way unique to conversations about mental illness that. Hm. Maybe isn't great? That sometimes feels a little "gotcha"! to me. And I think that framing is part of why we keep trying to have a Conversation About Depression that is the exact same thing year after year while more and more people die and we're not worrying enough about the equivalent of a Cancer Moonshot or making that not happen in a practical sense.

*being a person in society is however it is, and a difficult bummer for a lot of people, and maybe here is where it's worth pointing out mental illness IS in large part a societal problem that is treated even by the most forward-thinking as a personal responsibility, but also that difficulty is not the same thing as clinical depression so can we not do that anymore too
posted by colorblock sock at 11:43 AM on May 19 [17 favorites]


Surprised to see zero acknowledgement of epigenetics in all this (essentially, alterations not in DNA sequence but in the way genes are switched on and off). It's outside the scope of this study's methodology, but it's an important contributor to the heritability of most diseases. Mendelian genetics and single nucleotide polymorphisms (SNPs) only impact relatively few monogenic forms of disease and depression ain't one.
posted by basalganglia at 11:48 AM on May 19 [5 favorites]


Also, please remember the difference between statistical significance and clinical significance. A p-value (was this finding due to random chance?) is not very helpful without an accompanying F-statistic or B-coefficient (what is the effect size of this intervention, or how much of the change in [variable] is due to this specific characteristic/intervention/drug?) This is a particular problem when your sample size goes up; something that is "significant at p<0.05" might only extend life by 3 days, or might only prevent 1 heart attack for every 10k people, or might reduce your PHQ-9 score by 1 point.
posted by basalganglia at 12:00 PM on May 19 [7 favorites]


Maybe our genes have evolved not just to do their jobs, but to be robust to mutations, so that - with a few dramatic exceptions - if you swap out one amino acid for another, you still get a protein that works reasonably well?

That is exactly what happens. Many single-codon changes result in the same amino acids and many others result in amino acids that are functionally similar.
posted by sjswitzer at 12:05 PM on May 19 [1 favorite]


There have been times when my life would not have been possible without anti-depressant medication. Getting off anti-depressants can be quite difficult. I still take a small dose of zoloft daily; reducing it has been good, but the last step of eliminating is not so easy. It seems very modern to me; produce new stuff, use lots of it; have no idea what the end game is.
posted by theora55 at 12:05 PM on May 19 [2 favorites]


No gene for depression? Just as we thought, it's a moral failing. sarcasm, of course, but in these times, might as well label it.
posted by theora55 at 9:04 AM on May 19 [3 favorites −]


Sadly, moral failing is the first thing I thought when I saw this, as I've been so ingrained with that message from so many directions. Mind you, I don't hold that belief at all for other people, only for myself. Fortunately a degree in Social Work, and many years of working in social services have helped me to recognize/understand (to a certain degree) /empathize with others who have mental health issues. I hope someday that will continue to trickle in to my own psyche and erode the self-bias.

Thank you Theora55 for helping me to see how I was reading this post 'against myself'. Also, thank you colorblock sock, for sharing what you did. Your comments were insightful and well articulated, and (at the very least - for me) a voice I needed to hear.
posted by BigHeartedGuy at 12:06 PM on May 19 [10 favorites]


It's 2019. Have the press still not grown out of this childish "1 gene == 1 broad semantic human trait" nonsense? This is like publishing "STILL NO GENE FOR SINGING AMERICAN PIE LYRICS CORRECTLY ALL THE WAY THROUGH" or something. It's like asking if there's a "rum-soaked space hobo circuit" in everyone's computers, to make them display this post.

Genetics is currently hacking its way through a jungle of side-effects, combinatoric explosions of protein effects, and epigenetic factors and controls. There are far too few genes in the human genome for every little trait we can think of to map to an individual protein.

Also not all traits are selective, dammit.
posted by rum-soaked space hobo at 12:23 PM on May 19 [14 favorites]


No gene for depression? Just as we thought, it's a moral failing. sarcasm, of course, but in these times, might as well label it.

I hadn't thought of that as the alternative explanation - negative life events seem like they have pretty good scientific support - though from what people are saying it sounds like "depression is a moral failing" is a pretty common idea.

(I wonder if being told all your life that your sadness is a moral failing is itself a negative life event that's likely to cause depression...)
posted by clawsoon at 12:23 PM on May 19 [7 favorites]


rum-soaked space hobo: It's 2019. Have the press still not grown out of this childish "1 gene == 1 broad semantic human trait" nonsense?

Lemme see what a quick Google News search turns up:

Scientists Find Genetic Variants That Prevent Obesity, Diabetes

Matchmaking site for genes leads scientists to autism candidate

Scientists discover a gene that regulates napping during the day

Scientists find genes with large effects on head and brain size

Scientists Find A Gene That Could Lead to Erectile Dysfunction

Scientists find genetic variants that increase risk of ADHD

Scientists Look for the Genes That Determine Beauty

Scientists Find Genetic Causes of Loneliness

Some people have a gene linked with happy marriages, scientists find

All of these are from the past year.
posted by clawsoon at 12:35 PM on May 19 [3 favorites]


In mice.

(at best)
posted by sjswitzer at 12:49 PM on May 19 [5 favorites]


Happily married mice.
posted by clawsoon at 12:54 PM on May 19 [14 favorites]


Some mice have a gene linked with aspiring to pilot chefs around by their hair. It's science!
posted by nebulawindphone at 12:59 PM on May 19 [6 favorites]


Ed Yong has an article on the FPP study that I wish I would've found before I made the post. Good summary of the history and issues involved.
When he and others finally did a large study in 2005—with 100,000 people rather than the 1,000 from the original 1996 paper—they got nothing.

“You would have thought that would have dampened enthusiasm for that particular candidate gene, but not at all,” he says. “Any evidence that the results might not be reliable was simply not what many people wanted to hear.” In fact, the pace at which SLC6A4/depression papers were published accelerated after 2005, and the total number of such papers quadrupled over the next decade. “We’re told that science self-corrects, but what the candidate gene literature demonstrates is that it often self-corrects very slowly, and very wastefully, even when the writing has been on the wall for a very long time,” Munafo adds.

Many fields of science, from psychology to cancer biology, have been dealing with similar problems: Entire lines of research may be based on faulty results. The reasons for this so-called “reproducibility crisis” are manifold. Sometimes, researchers futz with their data until they get something interesting, or retrofit their questions to match their answers. Other times, they selectively publish positive results while sweeping negative ones under the rug, creating a false impression of building evidence.

Beyond a few cases of outright misconduct, these practices are rarely done to deceive. They’re an almost inevitable product of an academic world that rewards scientists, above all else, for publishing papers in high-profile journals—journals that prefer flashy studies that make new discoveries over duller ones that check existing work. People are rewarded for being productive rather than being right, for building ever upward instead of checking the foundations. These incentives allow weak studies to be published. And once enough have amassed, they create a collective perception of strength that can be hard to pierce.
posted by clawsoon at 12:59 PM on May 19 [7 favorites]


Also, please remember the difference between statistical significance and clinical significance. A p-value (was this finding due to random chance?) is not very helpful without an accompanying F-statistic or B-coefficient (what is the effect size of this intervention, or how much of the change in [variable] is due to this specific characteristic/intervention/drug?)

Agree completely with the gist of this comment. However, I came here as someone interested in causal inference to point out that an F-statistic is not a causal or effect-size measure. It's like a chi-squared or t-statistic. It's the thing the p-value is determined by. Odds ratios, risk ratios, rate ratios, incidence ratios, rate differences, ratios and differences of means, and beta coefficients in regressions most definitely are causal parameters when used to measure effect sizes.
posted by Mental Wimp at 2:38 PM on May 19 [3 favorites]


There definitely are depression jeans though and they are inevitably too tight.
posted by srboisvert at 2:51 PM on May 19 [5 favorites]


Thanks for the correction, Mental Wimp. My brain is broken from too much Stata this weekend!
posted by basalganglia at 3:35 PM on May 19


I hadn't thought of that as the alternative explanation - negative life events seem like they have pretty good scientific support - though from what people are saying it sounds like "depression is a moral failing" is a pretty common idea.

People with depression (and most mood disorders, and pretty much everything else that falls under the umbrella of behavioral health) are pretty consistently given the message that the reason they are suffering is a failure to want not to suffer. Cf: those fucking "just go outside!" FB posts
posted by PMdixon at 5:38 PM on May 19


BigHeartedGuy, clawsoon, I have been told all my life to Try Harder. By family, teachers, ex-husband, myself. Turns out, I have ADD, seem to be pre-disposed to depression, and my serious hearing issues were not diagnosed until I was 40. I have been Trying Hard, and mostly coping, my whole life, while dealing with some big hurdles. Many people have hurdles, but a lot of us have unseen hurdles. Be kind, for you don't know what burden another person may be carrying. They may not even know.
posted by theora55 at 5:39 PM on May 19 [21 favorites]


In a perfect world, people who sometimes experience melancholy (genetic or otherwise?) can contribute unique insights and perspectives that can help to make society as a whole more healthy, interesting, & diverse.

(In an imperfect world that doesn't have universal socialized medicine, the corporate health insurance industry will just pick away at your DNA profile to find reasons to deny health care);
posted by ovvl at 5:47 PM on May 19


Those meta studies don’t seem to have included lithium (unless I missed it). I’ve been fascinated with lithium as a drug based on its biochemistry and pharmacodynamics — it’s similar to sodium, but smaller! — ever since my late mother was prescribed it for treatment of manic-depressive (and later, bipolar) disorder. I recall it took a lot of fiddling with the dosing to dial it in, and then it just seemed to stop working for her.

I’ve read a fair bit on lithium, and Wikipedia covers it seemingly quite well, but now I’m thinking what I read may suffer from the same flaws as those suggested by folks in this thread. Does anyone here know of the latest meta-research on lithium (as the carbonate or chloride) as an antidepressant?
posted by darkstar at 9:15 PM on May 19 [1 favorite]


Google brings up this meta-analysis on lithium, which is generally positive. But I’d be grateful for any insights on the study from someone who is rather more stats-savvy than I am.
posted by darkstar at 9:59 PM on May 19


People with depression (and most mood disorders, and pretty much everything else that falls under the umbrella of behavioral health) are pretty consistently given the message that the reason they are suffering is a failure to want not to suffer.

Discarding the idea that depression and mental disorders have an organic cause in favor of "they choose to not be happy", seems to be the perfect match to a society that is in the process of eliminating any sense of empathy.

This isn't so far from "Poor people just have bad financial habits! Avocado toast!". And it stems from the same impulse- condemn rather than help.
posted by happyroach at 10:00 PM on May 19


I wonder what, 500 years from now when science has presumably figured out cognition and genetics, people will think of our current search for “_ genes.” To me it seems like obvious reductionism. People are complicated, and so is depression. There isn’t going to be a cute explanation or a universal cure. But that doesn’t mean there isn’t a biological basis or that it’s impossible to understand.
posted by mantecol at 10:44 PM on May 19


why did Johnson et al. bother to look at schizophrenia candidate genes and then exclude the recently discovered C4 variants in the MHC region? C4 has an incredibly strong link to SCZ...

" we present evidence that several of the most-studied candidate genes—particularly NOTCH4, DRD2, KCNN3, GRM3, and TNF—are more strongly related to schizophrenia than expected by chance. It is important to put this evidence in perspective. First, two of these five genes (NOTCH4 and TNF) are in the MHC, and given the long-range and complex nature of LD in this region, it is unclear whether it is variants in these two genes or variants in some of the other ~240 genes in the MHC that are relevant, and indeed recent evidence suggests the signal driving the NOTCH4 association comes not from NOTCH4 but from the nearby complement component 4 (C4A and C4B) MHC genes (25.[ Sekar et al. previously ]). For this reason, we hereafter restrict discussion of the most significantly associated candidate genes to those not in the MHC. "
posted by dongolier at 12:23 AM on May 20


I had forgotten that I had posted that link, dongolier. Interesting. Maybe they excluded that gene because it was found using a sufficiently-powered GWAS study (100,000 people) instead of the "candidate gene" approach that they're criticizing?
posted by clawsoon at 4:19 AM on May 20


More from the schizophrenia paper:
Additionally, the imputed data used by the PGC GWAS analysis did not adequately capture common polymorphisms within all the candidate genes. For example, the most frequently studied candidate polymorphisms for DRD4 and NOTCH4 were neither genotyped nor successfully imputed in the PGC GWAS. The most commonly studied polymorphism in DRD4 is a 13-bp indel, which is not well captured in the kind of sequencing done by the 1000 Genomes(21) project, the reference panel used to impute in the PGC data. The most commonly studied polymorphism in NOTCH4, rs367398, was also missing from the 1000 Genomes phase 1 reference panel, and the PGC data included no SNPs in LD with that polymorphism at R2 > 0.3. As already noted, this SNP is located in the MHC region, which has made genotyping and analyzing this section of the genome difficult, although a recent study suggests that a large proportion of the genetic risk to schizophrenia in the MHC is specific to a particular locus in the C4A gene (25).
So perhaps the title of the paper should've come with an asterisk: "No Evidence That Schizophrenia Candidate Genes Are More Associated With Schizophrenia Than Noncandidate Genes*"

"*except maybe one of them"

One of the authors has a couple of comments in the SSC discussion that help make sense of the paragraph I quoted here. In particular:
Due to the autocorrelated nature of the genome (called linkage disequilibrium), with just 800k SNPs on an array (or even 250k), one can predict virtually all of the other common variants.
So, if I'm understanding this correctly: If you pick the right collection of cheap-to-find single-bit-flip genetic variations - cheap-to-find being what allows you to build a database of 100,000 individuals - you can safely make assumptions about almost all of the other genetic variations that someone has. However, you can't make assumptions about absolutely all of them using this method, and a couple of genes you can't make assumptions about are schizophrenia candidate genes.
posted by clawsoon at 4:56 AM on May 20


People with depression (and most mood disorders, and pretty much everything else that falls under the umbrella of behavioral health) are pretty consistently given the message that the reason they are suffering is a failure to want not to suffer.

I actually have not encountered this at all. Maybe I'm just lucky. But in 20 years of dealing with anxiety and depression (as well as behavioral issues with my kid) I've found professionals to be well, professional, about helping. It's frustrating that there aren't better and more easily accessible treatments, but I've never had anyone tell me my issues are a character flaw.
posted by schwinggg! at 6:40 AM on May 20


One of the things that's maddening about biases in mental health care is that they're different depending on your demographic — so you can't count on your own experience to tell you what biases other people face.

As a white woman in tech, my experience is that antidepressants and ADHD meds are if anything overprescribed in my social circle. But if you're a young man of color who's seen as a likely criminal you probably can't get stimulants, and if you're a black mom who's expected to be Stoic and Long-Suffering you probably have a hard time getting either.

Or, like, as an adult with money and a stable life, my experience was that getting on antipsychotics was way too hard — my doctors kept me on SSRIs for too long, and needed a lot of evidence they weren't working before changing the plan. But if you're a trauma survivor with dissociative symptoms and a history of homelessness, you can get thrown on antipsychotics for very little reason. Similarly if you're a poor kid who's coded as Noncompliant and your school gets its special needs funding in ways that incentivize it.

(I'm picking dramatic examples. But I also know other rich white women with professional jobs who trigger different provider biases because of smallish differences in personality or life circumstance, like "Are they afraid of doctors?" or "What ways do they mask their symptoms?")

And so similarly the messages we get aren't the same. I definitely get the message from my providers and my friends that I deserve self-care and support, and that being on antidepressants is basically a spa day for my neurotransmitters. (Though now that I've been in the hospital a few times and I'm on an antipsychotic, not everyone sends that message anymore...) But other people do get the message that their mental illness is pure laziness, or a moral failing, or a thing they're making up for attention, or proof that they're one of Those People, or a lie invented by people in power just to disempower them, or etc.

So talking about What Our Culture Believes about mental illness is oversimplifying, and you end up in this sort of "WTF where are you even getting that idea?" situation.
posted by nebulawindphone at 7:06 AM on May 20 [12 favorites]


One of the authors (Matthew C. Keller) corrected the framing of my post in email, and graciously allowed me to quote him directly here:
We are not saying that there are no genes that influence depression, nor even that science hasn't done a good job at already having identifying several of them (via GWAS). It's just that one particular and increasingly outdated approach to finding them, called the "candidate gene approach", has been an abysmal failure. The candidate gene approach was the primary approach in the 90s and 2000s, and much of what you see in the popular press, wiki, etc are from that era - and it's basically all wrong.

Compare that to the modern approach - GWAS - where the effect sizes are tiny, the samples huge, and the results highly replicable (which we demonstrated in our paper).
Credit to pharm, who made part of this point upthread.
posted by clawsoon at 9:00 AM on May 20 [10 favorites]


yeah, clawsoon that is exactly the lesson from the Sekar et al. (Nature 2016) schizophrenia work: they had to build a model based on the genetic sequence they were looking at, it had stuff like human retrovirus sequences right in the middle of the C4A and C4B alleles ...a naive statistical GWAS approach gave only a vague schizophrenia link to 'somewhere in the the MHC area' of the genome (later mis-attributed to NOTCH4, apparently).

The beautiful thing was when the Broad Institute team really got their hands dirty the story unfolded...two very similar C4 proteins, only a few bases were different at the active binding location, one behaved like a base the other like an acid ...and if that difference feeds a living immune process of synaptic pruning in late adolescence...well that is a real insight into how human brains might really be very different from one another starting from a very subtle genetic nudge. Add the 'slow prion' work re: Alzheimers a completely new direction for what has been a fruitless genetic search, and you are led to conclude that meany people think genetics should have all the answers and the work should be simple, straightforward and easy ...its just statistics after all?

No, its actually random mutations and natural selection applied over the last two billion years, there's definitely a story there, but the only forensic evidence is a string of three billion letters.... ATTAGCCAT....
posted by dongolier at 3:04 AM on May 21 [1 favorite]


dongolier: ...a naive statistical GWAS approach gave only a vague schizophrenia link to 'somewhere in the the MHC area' of the genome (later mis-attributed to NOTCH4, apparently).

My impression is that most of our genome is pretty predictable, so that if you have mutations A, B and C we can be pretty confident that you also have D, but that's not the case in the MHC region. Figuring out linkage disequilibrium for the MHC, so that GWAS can be used with it, appears to be a very active area of study.
posted by clawsoon at 6:12 AM on May 21




« Older Jon Batiste's 'The Game of Thrones' Song   |   The Social Dynamics of Fantasy Dragons Newer »


This thread has been archived and is closed to new comments