Thoughts of Blue Brains and GABA Interneurons

An unsuccessful plan to create a computer simulation of a human brain within 10 years. An exhaustive catalog of cell types comprising a specific class of inhibitory neurons within mouse visual cortex. What do these massive research programs have in common? Both efforts were conducted by large multidisciplinary teams at non-traditional research institutions: the Blue Brain Project based in Lausanne, Switzerland and the Allen Institute for Brain Science in Seattle, Washington.

BIG SCIENCE is the wave of the future, and the future is now. Actually, that future started 15-20 years ago. The question should be, is there a future for any other kind of neuroscience?
 

Despite a superficial “BIG SCIENCE” similarity, the differences between funding sources, business models, leadership, operation, and goals of Blue Brain and the Allen Institute are substantial. Henry Markram, the “charismatic but divisive” visionary behind Blue Brain (and the €1 billion Human Brain Project) has been criticized for his “autocratic” leadership, “crap” ideas, and “ill-conceived, ... idiosyncratic approach to brain simulation” in countless articles. His ambition is undeniable, however:

“I realized I could be doing this [eg., standard research on spike-timing-dependent plasticity] for the next 25, 30 years of my career, and it was still not going to help me understand how the brain works.”

 

I'm certainly not a brilliant neuroscientist in Markram's league, but I commented previously on how a quest to discover “how the brain works” might be futile:

...the search for the Holy Grail of [spike trains, network generative models, manipulated neural circuit function, My Own Private Connectome, predictive coding, the free energy principle (PDF), or a computer simulation of the human brain promised by the Blue Brain Project] that will “explain” how “The Brain” works is a quixotic quest. It's a misguided effort when the goal is framed so simplistically (or monolithically).

In his infamous 2009 TED talk, Markram stated that a computer simulation of the human brain was possible in 10 years:

“I hope that you are at least partly convinced that it is not impossible to build a brain. We can do it within 10 years, and if we do succeed, we will send to TED, in 10 years, a hologram to talk to you.”

This claim would come back to haunt him in 2019, because (of course) he was nowhere close to simulating a human brain. In his defense, Markram said that his critics misunderstood and misinterpreted his grandiose proclamations.¹

Blue Brain is now aimed at “biologically detailed digital reconstructions and simulations of the mouse brain.”

In Silico

Documentary filmmaker Noah Hutton² undertook his own 10 year project that followed Markram and colleagues as they worked towards the goals of Blue Brain. He was motivated by that TED talk and its enthralling prediction of a brain in a supercomputer (hence in silico). Originally entitled Bluebrain and focused on Markram, the documentary evolved over time to include more realistic viewpoints and interviews with skeptical scientists, including Anne Churchland, Terry Sejnowski, Stanislas Dehaene, and Cori Bargmann. Ironically, Sebastian Seung was one of the loudest critics (ironic because Seung has a grandiose TED talk of his own, I Am My Connectome).

 

In Silico was available for streaming during the DOC NYC Festival in November (in the US only), and I had the opportunity to watch it. I was impressed by the motivation and dedication required to complete such a lengthy project.  Hutton had gathered so much footage that he could have made multiple movies from different perspectives.

Over the course of the film, Blue Brain/Human Brain blew up, with ample critiques and a signed petition from hundreds of neuroscientists (see archived Open Letter).

And Hutton grew up. He reflects on the process (and how he changed) at the end of film. He was only 22 at the start, and 10 years is a long time at any age.

Some of the Big Questions in In Silico:

• How do you make sure all this lovely simulated activity would be relevant for an animal's behavior?

• How do you build in biological imperfections (noise) or introduce chaos into your perfect pristine computational model? “Tiny mistakes” are critical for adaptable biological systems.

• “You cannot play the same soccer game again,” said one of the critics (Terry Sejnowski, I think)

• “What is a generic brain?”

• What is the vision? 

The timeline kept drifting further and further into the future. It was 10 years in 2009, 10 years in 2012, 10 years in 2013, etc. 

Geneva 2019, and it's Year 10 – only two Principals left, 150 papers published, and a model of 10 million neurons in mouse cortex. Stunning visuals, but still disconnected from behavior.

In the end, “What have we learned about the brain? Not much. The model is incomprehensible,” to paraphrase Sejnowski.

GABA Interneurons

Another brilliant and charismatic neuroscientist, Christof Koch, was interviewed by Hutton. “Henry has two personalities. One is a fantastic, sober scientist … the other is a PR-minded messiah.”

Koch is Chief Scientist of the MindScope Program at the Allen Institute for Brain Science, which focuses on how neural circuits produce vision. Another major unit is the Cell Types Program, which (as advertised) focuses on brain cell types and connectivity.³

The Allen Institute core principles are team science, Big Science, and open science. An impressive recent paper by Gouwens and 97 colleagues (2020) is a prime example of all three. Meticulous analyses of structural, physiological, and genetic properties identified 28 “met-types” of GABAergic interneurons that have congruent morphological, electrophysiological, and transcriptomic properties. This was winnowed down from more than 500 morphologies in 4,200 GABA-containing interneurons in mouse visual cortex. With this mind-boggling level of neuronal complexity in one specific class of cells in mouse cortex — along with the impossibility of “mind uploading” — my inclination is to say that we will never (never say never) be able to build a realistic computer simulation of the human brain.

Gouwens et al., 2020

Footnotes 

¹ Here's another gem: “There literally are only a handful of equations that you need to simulate the activity of the neocortex.”

² Most of Hutton's work has been as writer and director of documentary films, but I was excited to see that his first narrative feature, Lapsis, will be available for streaming next month. To accompany his film, he's created an immersive online world of interlinked websites that advertise non-existent employment opportunities, entertainment ventures, diseases, and treatments. It very much reminds me of the realistic yet spoof websites associated with the films Eternal Sunshine of the Spotless Mind (LACUNA, Inc.) and Ex Machina (BlueBook). In fact, I'm so enamored with them that they've appeared in several of my own blog posts.

³ Investigation of cell types is big in the NIH BRAIN Initiative ® as well.



References

Abbott A. (2020). Documentary follows implosion of billion-euro brain project. Nature 588:215-6.

[Alison Abbott covered the Blue Brain/Human Brain sturm und drang for years]

Gouwens NW, Sorensen SA, Baftizadeh F, Budzillo A, Lee BR, Jarsky T, Alfiler L, Baker K, Barkan E, Berry K, Bertagnolli D ... Zeng H et al. (2020). Integrated morphoelectric and transcriptomic classification of cortical GABAergic cells. Cell 83(4):935-53.

Waldrop M. (2012). Computer modelling: Brain in a box. Nature News 482(7386):456.

Further Reading

The Blue Brain Project (01 February 2006), by Dr. Henry Markram

“Alan Turing (1912–1954) started off by wanting to 'build the brain' and ended up with a computer. ... As calculation speeds approach and go beyond the petaFLOPS range, it is becoming feasible to make the next series of quantum leaps to simulating networks of neurons, brain regions and, eventually, the whole brain.”

A brain in a supercomputer (July 2009), Henry Markram's TED talk

• from the transcript:

“Our mission is to build a detailed, realistic computer model of the human brain. And we've done, in the past four years, a proof of concept on a small part of the rodent brain, and with this proof of concept we are now scaling the project up to reach the human brain.”

Blue Brain Founder Responds to Critics, Clarifies His Goals (11 Feb 2011), Science news

Bluebrain: Noah Hutton's 10-Year Documentary about the Mission to Reverse Engineer the Human Brain (9 Nov 2012), an indispensable interview with Ferris Jabr in Scientific American

European neuroscientists revolt against the E.U.'s Human Brain Project (11 July 2014), Science news

Row hits flagship brain plan (7 July 2014), Nature news

Brain Fog (7 July 2014), Nature editorial

Human Brain Project votes for leadership change (4 March 2015), Nature news

'In Silico:' Director Noah Hutton reveals how one neuroscientist's pursuit of perfection went awry (10 Nov 2020), another indispensable interview, this time with Nadja Sayej in Inverse

“They still haven’t even simulated a whole mouse brain. I realized halfway through the 10-year point that the human brain probably wasn’t going to happen.” ...

In the first few years, I followed only the team. Then, I started talking to critics.

 

 

Brain in a box

The Future of Depression Treatment

2014

Jessica is depressed again. After six straight weeks of overtime, her boss blandly praised her teamwork at the product launch party. And the following week she was passed over for a promotion in favor of Jason, her junior co-worker. "It's always that way, I'll never get ahead..."

She arrives at her therapist's office late, looking stressed, disheveled, and dejected. The same old feelings of worthlessness and despair prompted her to resume her medication and CBT routine.

"You deserve to be recognized for your work," said Dr. Harrison. "The things you're telling yourself right now are cognitive distortions: the black and white thinking, the overgeneralization, the self-blame, jumping to conclusions... " 

"I guess so," muttered Jessica, looking down.

"And you need a vacation!"

. . .

A brilliant suggestion, Dr. Harrison. As we all know, taking time off to relax and recharge after a stressful time will do wonders for our mental health. And building up a reserve of happy memories to draw upon during darker times is a cornerstone of positive psychology.

Jessica and her husband Michael take a week-long vacation in Hawaii, creating new episodic memories that involve snorkling, parasailing, luaus, and mai tais on the beach. Jessica ultimately decides to quit her job and sell jewelry on Etsy.

2015

Michael is depressed after losing his job. His self-esteem has plummeted, and he feels useless. But he's too proud to ask for help. "Depression is something that happens to other people (like my wife), but not to me." He grows increasingly angry and starts drinking too much.

Jessica finally convinces him to see Dr. Harrison's colleague. Dr. Roberts is a psychiatrist with a Ph.D. in neuroscience. She's adopted a translational approach and tries to incorporate the latest preclinical research into her practice. She's intrigued by the latest finding from Tonegawa's lab, which suggests that the reactivation of a happy memory is more effective in alleviating depression than experiencing a similar event in the present.

Recalling happier memories can reverse depression, said the MIT press release. 

So instead of telling Michael to take time off and travel and practice mindfulness and live in the present, she tells him to recall his fondest memory from last year's vacation in Hawaii.  

It doesn't work.

Michael goes to see Dr. Harrison, who prescribes bupropion and venlafaxine. Four weeks later, he feels much better, and starts a popular website that repudiates positive psychology. Seligman and Zimbardo are secretly chagrined. 

. . .

Happy Hippocampus

photo credit: S. Ramirez

Artificially reactivating positive [sexual] memories [in male mice] could offer an alternative to traditional antidepressants makes them struggle more when you hold them by the tail after 10 days of confinement.¹

Not as upbeat as the press release, eh?

The findings ... offer a possible explanation for the success of psychotherapies in which depression patients are encouraged to recall pleasant experiences. They also suggest new ways to treat depression by manipulating the brain cells where memories are stored...

“Once you identify specific sites in the memory circuit which are not functioning well, or whose boosting will bring a beneficial consequence, there is a possibility of inventing new medical technology where the improvement will be targeted to the specific part of the circuit, rather than administering a drug and letting that drug function everywhere in the brain,” says Susumu Tonegawa, ... senior author of the paper.

Although this type of intervention is not yet possible in humans, “This type of analysis gives information as to where to target specific disorders,” Tonegawa adds.

Before considering what the mice might actually experience when their happy memory cells are activated with light, let's all marvel at what was accomplished here.

Ramirez et al. (2015) studied mice that were genetically engineered to allow blue light to activate a specific set of granule cells in the dentate gyrus subfield of the hippocampus. These neurons are critical for the formation of new memories and are considered “engram cells” that undergo physical changes and store discrete memories (Liu et al., 2014). When a cue reactivates the same set of neurons, the episodic memory is retrieved. In this study, the engram cells were part of a larger circuit that included the amygdala and the nucleus accumbens, regions important for processing emotion, motivation, and reward.

Ramiriez, Liu, Tonegawa and colleagues have repeatedly demonstrated their masterful manipulation of mouse memories: activating fear memories, implanting false memories, and changing the valence of memories. These experiments are technically challenging and far outside my areas of expertise (greater detail in the Appendix below). In brief, the authors were able to label discrete sets of dentate gyrus cells while they were naturally activated during an interval of positive, neutral, or negative treatment. Then some groups of  animals were stressed for 10 days, and others remained in their home cages.

The stressed mice exhibited signs of “depression-like” and “anxiety-like” behaviors.²  I'll spare you the long digression about whether the tail suspension test successfully models the anguished human experience of abject states, but you can read my earlier musings on the topic.


The most astounding part of the experiment is that optical stimulation of positive-memory engram cells in stressed mice induced a reversal of “depressive” behaviors (but not “anxious” behaviors; see Appendix). Curiously, re-exposing the stressed male mice to an actual female did not have this positive benefit. So mediated experience — artificial reactivation of the engram — is even better than the real thing.

The first author, graduate student Steve Ramirez, offered a post hoc explanation:

“People who suffer from depression have those positive experiences in the brain, but the brain pieces necessary to recall them are broken. What we’re doing, in mice, is bypassing that circuitry and forcing it to be jump-started,” Ramirez says. “We’re harnessing the brain’s power from within itself and forcing the activation of that positive memory, whereas if you give a natural positive memory to the person or the animal, the depression that they have prevents them from finding that experience rewarding.”

In other words, “We'll force you to be happy [i.e., possibly remember a positive experience], whether you like it or not.” And since the authors discussed therapeutic implications in the paper, they have to deal with the problem of phenomenology, whether they like it or not. What do the mice actually remember? Generic sexual experiences, a feeling of reward? An episodic-like memory, e.g. a specific act and all its spatiotemporal contextual information? Even if we allow mice to have “episodic-like” memories, the latter seems unlikely given the highly artificial and non-physiological method of neural stimulation that bypasses the precisely timed patterns of activity thought to “represent” past experience. These memory manipulation studies seem very futuristic and scary but Inception they are not.

Our memories are plastic and malleable, and their physical instantiation changes each time we recall them. Which version of the Hawaii trip shall we target? What other memories show the greatest overlap with the happy one? Has the problem of hippocampal pattern separation been solved already?? Garden-variety deep brain stimulation seems easy in comparison (and we know how well that's gone in humans, so far). But: “In rodents, optogenetic stimulation of mPFC neurons, mPFC to raphe projections, and ventral tegmental dopaminergic neurons achieved a rapid reversal of stress-induced maladaptive behaviours” (Ramirez et al., 2015).

Why can't we just appreciate the basic knowledge gained from these experiments? But no. There has to be a human application right around the corner.

That link between the neural circuit manipulations in mice and therapies now used in humans makes the findings particularly exciting, says Tom Insel, director of the National Institute of Mental Health.

“This is a big step toward helping to understand not only the underlying circuits for a really serious illness like depression, but also the circuits that underlie treatment,” says Insel...

Was that actually an endorsement of mediated experience? If we go down that road, we must acknowledge that an artificially created reality, albeit one that originates within a being's own brain, is superior to real life. This is the most profound implication of activating positive memory engrams.


When Mediated Experience Replaces a Medicated Existence

Mediated experiences increasingly dominate our lives. Movies and television already confuse the real and the mediated. New technology is blurring the line further. Video games and virtual reality are becoming increasingly realistic. “Augmented reality” technology is on its way to the public. Wearable computers will allow people to enter a news story and see and feel the events the way the journalist who was there did and no doubt eventually we’ll be able to experience the events live. As the line between real and mediated gets harder to see, presence increases. An important and overlooked consequence of this trend is an increasing confusion from the other direction, in which “real life” seems to be mediated. People will have more and more trouble distinguishing reality, and some may not even appreciate that there is a difference. It will get harder for people to trust their own senses and judgment and it will be more difficult to impress people with non-mediated experiences.

Reeves Timmins & Lombard (2005).  When “Real” Seems Mediated: Inverse Presence.

Heavy social media users already accept a reality filtered through Instagram and Facebook. As the interest in personal biometrics and the Quantified Self movement rises, so too will tolerance of increasingly invasive performance enhancing and “lifestyle” brain stimulation methods (see DIY tDCS). No one has said that optogenetic-type treatments are (or will be) possible in humans (OK, almost no one; see Albert, 2014). Others are more modest, and see the translational potential in non-invasive transcranial magnetic stimulation (Deisseroth et al., 2015).

. . .

2035

DARPA has mandated that all depressed Americans must be implanted with its CyberNeuroTron WritBit device, which cost $100 billion to develop. CNTWB is a closed-loop DBS system that automatically adjusts the stimulation parameters at 12 different customized target locations. It uses state-of-the-art syringe-injectable mesh electronics, incorporating silicon nanowires and microvoltammetry. Electrical and chemical signals are continuously recorded and uploaded to a centralized data center, where machine learning algorithms determine with high accuracy whether a given pattern of activity signals a significant change in mood.

The data are compiled, analyzed, and stored by the global search engine conglomerate BlueBook, which in 2032 swallowed up Google, Facebook, Apple, and every other internet data mining company.

. . .

2055

Sophia, the daughter of Jessica and Michael, is depressed again. The Ramirez et al. (2050) protocol for Positive Memory Engram Activation is in widespread use. Sophia searches for her dentate gyrus recordings from a vacation in Hawaii five months earlier. Then she selects the specific memory she wants to be artificially reactivated: watching the sunset on the beach with her partner, drinking mai tais and eating taro chips.

"We had a great time on that trip, didn't we Lucas?" 

Lucas the intelligent AI nods in agreement. "It's true," he thought. "Humans can no longer distinguish between virtual reality and the real thing."

This has been especially useful for the Ramirez protocol, since most Pacific Island nations have been underwater since 2047.



Footnotes

¹ As an aside, I wonder what the female mice think of all this. What would be an equivalently positive experience? Is sex as rewarding for them? Will there be a new animal model of shopping at Nordstrom? Fortunately, this work was funded by RIKEN Brain Science Institute and Howard Hughes Medical Institute, so the authors don't have to follow the pesky impending NIH guidelines to include females in animal research.

² “Depression-related” behaviors were assessed using the Tail Suspension Test (TST) and the Sucrose Preference Test (SPT), which are supposed to mimic giving up hope and loss of pleasure, respectively. Different tests were used to measure “anxiety-related” behaviors. Interestingly, none of the happy engram manipulations improved anxiety-like behavior in the mice. Not a very good model of anxious depression, then.


References

Albert PR. (2014). Light up your life: optogenetics for depression? J Psychiatry Neurosci. 39(1):3-5.

Deisseroth K, Etkin A, Malenka RC. (2015). Optogenetics and the circuit dynamics ofpsychiatric disease. JAMA 313(20):2019-20.

Liu, X., Ramirez, S., Redondo, R., & Tonegawa, S. (2014). Identification and Manipulation of Memory Engram Cells Cold Spring Harbor Symposia on Quantitative Biology, 79, 59-65. DOI: 10.1101/sqb.2014.79.024901

Ramirez, S., Liu, X., MacDonald, C., Moffa, A., Zhou, J., Redondo, R., & Tonegawa, S. (2015). Activating positive memory engrams suppresses depression-like behaviour. Nature, 522 (7556), 335-339. DOI: 10.1038/nature14514

Timmins, L., & Lombard, M. (2005). When “Real” Seems Mediated: Inverse Presence. Presence: Teleoperators and Virtual Environments, 14 (4), 492-500. DOI: 10.1162/105474605774785307


Appendix

These experiments are indeed difficult, but if you successfully execute them, a publication is Nature nearly guaranteed. A review by Liu et al. (2014) explained their general protocol in an easier-to-understand fashion:

...we combined activity-dependent, drug-regulatable expression system with optogenetics (Liu et al. 2012). We used a transgenic mouse model where the artificial tetracycline transactivator (tTA), which can be blocked by doxycycline (Dox), is driven by the promoter of immediate early gene (IEG) c-fos (Reijmers et al. 2007). The activity dependency of c-fos promoter poses a natural spatial constrain on the identities of the neurons that can be labeled, reflecting the normal biological selection process of the brain during memory formation, whereas the Dox-dependency of the system poses an artificial temporal constrain as to when these neurons can be labeled, which can be controlled by the experimenters. With these two constraints, the down-stream effector of tTA can express selectively in neurons that are active during a particular behavior episode, only if the animals are off Dox diet. Using this system, we expressed channelrhodopsin-2 (ChR2) delivered by a viral vector AAV-TRE-ChR2-EYFP targeting the dentate gyrus (DG) of the hippocampus and implanted optical fibers right above the infected areas. 

One of the major treatment protocols is shown below (adapted from Fig. 1A).

There were a number of control conditions too. Reactivation of neutral or negative engram neurons didn't change depression-like behaviors on the TST and SPT.  Reactivation of positive engram neurons in non-stressed mice didn't alter behavior, either.

A very impressive body of work, with a special dedication by the authors: "We dedicate this study to the memory of Xu Liu, who made major contributions to memory engram research."

Xu Liu in memoriam.