Behold, The Blue Brain

The Blue Brain project releases their first major paper today and boy, it’s a doozy. Including supplements, it’s over 100 pages long, including 40 figures and 6 tables. In order to properly understand everything in the paper, you have to go back and read a bunch of other papers they have released over their years that detail their methods. This is not a scientific paper: it’s a goddamn philosophical treatise on The Nature of Neural Reconstruction.

The Blue Brain Project – or should I say Henry Markram? it is hard to say where the two diverge – aims to simulate absolutely everything in a complete mammalian brain. Except right now it sits at middle-ground: other simulations have replicated more neurons (Izhikevich had a model with 10^11 neurons of 21 subtypes). At the other extreme, MCell has completely reconstructed everything about a single neuron – down to the diffusion of single atoms – in a way that Blue Brain does not.

The focus of Blue Brain right now is a certain level of simulation that derives from a particular mindset in neuroscience. You see, people in neuroscience work at all levels: from the individual molecules to flickering ion channels to single neurons up to networks and then whole brain regions. Markram came out of Bert Sakmann’s lab (where he discovered STDP) and has his eye on the ‘classical’ tradition that stretches back to Hodgkin and Huxley. He is measuring distributions of ion channels and spiking patterns and extending the basic Hodgkin-Huxley model into tinier compartments and ever more fractal branching patterns. In a sense, this is swimming against the headwinds of contemporary neuroscience. While plenty of people are still doing single-cell physiology, new tools that allow imaging of many neurons simultaneously in behaving animals have reshaped the direction of the field – and what we can understand about neural networks.

Some very deep questions arise here: is this enough? What will this tell us and what can it not tell us? What do we mean when we say we want to simulate the brain? How much is enough? We don’t really know – though the answer to the first question is assuredly no – and we assuredly don’t know enough to even begin to answer the second set of questions.


The function of the new paper is to collate in one place all of the data that they have been collecting – and it is a doozy. They report having recorded and labeled >14,000 (!!!!!) neurons from somatosensory cortex of P14 rats with full reconstruction of more than 1,000 of these neurons. That’s, uh, a lot. And they use a somewhat-convoluted terminology to describe all of these, throwing around terms like ‘m-type’ and ‘e-type’ and ‘me-type’ in order to classify the neurons. It’s something, I guess.


Since the neurons were taken from different animals at different times, they do a lot of inference to determine connectivity, ion channel conductance, etc. And that’s a big worry because – how many parameters are being fit here? How many channels are being missed? You get funny sentences in the paper like:

[We compared] in silico (edmodeled) PSPs with the corresponding in vitro (ed – measured in a slice prep) PSPs. The in silico PSPs were systematically lower (ed– our model was systematically different from the data). The results suggested that reported conductances are about 3-fold too low for excitatory connections, and 2-fold too low for inhibitory connections.

And this worries me a bit; are they not trusting their own measurements when it suits them? Perhaps someone who reads the paper more closely can clarify these points.

They then proceed to run these simulated neurons and perform ‘in silico experiments’. They first describe lowering the extracellular calcium level and finding that the network transitions from a regularly spiking state to a variable (asynchronous) state. And then they go, and do this experiment on biological neurons and get the same thing! That is a nice win for the model; they made a prediction and validated it.

On the other hand you get statements like the following:

We then used the virtual slice to explore the behavior of the microcircuitry for a wide range of tonic depolarization and Ca2+ levels. We found a spectrum of network states ranging from one extreme, where neuronal activity was largely synchronous, to another, where it was largely asynchronous. The spectrum was observed in virtual slices, constructed from all 35 individual instantiations of the microcircuit  and all seven instantiations of the average microcircuit.

In other words, it sounds like they might be able to find everything in their model.

But on the other hand…! They fix their virtual networks and ask: do we see specific changes in our network that experiments have seen in the past? And yes, generally they do. Are we allowed to wonder how many of these experiments and predictions did they do that did not pan out? It would have been great to see a full-blown failure to understand where the model still needs to be improved.

I don’t want to understand the sheer amount of work that was done here, or the wonderful collection of data that they now have available. The models that they make will be (already are?) available for anyone to download and this is going to be an invaluable resource. This is a major paper, and rightly so.

On the other hand – what did I learn from this paper? I’m not sure. The network wasn’t really doing anything, it just kind of…spiked. It wasn’t processing structured information like an animal’s brain would, it was just kind of sitting there, occasionally having an epileptic fit (note that at one point they do simulate thalamic input into the model, which I found to be quite interesting).

This project has metamorphosed into a bit of a social conundrum for the field. Clearly, people are fascinated – I had three different people send me this paper prior to its publication, and a lot of others were quite excited and wanted access to it right away. And the broader Blue Brain Project has had a somewhat unhappy political history. A lot of people – like me! – are strong believers in computation and modeling, and would really like it see it succeed. Yet what the chunk of neuroscience that they have bitten off, and the way they have gone about it, lead many to worry. The field had been holding its breath a bit to see what Blue Brain was going to release – and I think they will need to hold their breath a bit longer.


Markram et al (2015). Reconstruction and Simulation of Neocortical Microcircuitry Cell


24 thoughts on “Behold, The Blue Brain

  1. One thing that there’s to learn is that the bulk of published in vitro studies could be inferred without killing mice just by modeling them.

    You could direct the criticism the other way around: Computational neuroscience is doing the most it can given the data that is available, and the truth is the data is not really available. Most scientists publish behind paywalls and never make their data available. With such underconstrained models it’s not surprising that we can’t model a brain region properly. Part of it has to do with the focus of grant givers on funding small-scale studies and the ease with which these become published in reputable journals. It’s like trying to create a model of the weather while having only individual pictures of small clouds.

      • In fact it is published as a “Resource” — not as a research article or a methods paper. The issue remains that without a scientific result it is unclear if this resource is useful for the community at large.

  2. Anyone who uses the word ‘doozy’ in response to science cannot possibly be a scientist himself, or credible at all. Who are you? What have you achieved? We can all be pretty sure your ‘doozy’ scientific life will not amass to anything as groundbreaking as this paper. With an h-index of 3, I would be careful of attacking people who have contributed more significantly to the scientific world than yourself. So, do yourself a favour and shut up or get a more educated opinion so that you don’t end up coming across as a sour little boy with a ‘doozy’ attitude.

    • doozy, def: something outstanding or unique of its kind.

      I’m sorry that you have a problem with my using the word doozy. I like the word! It’s fun to say. Doozy doozy doozy. Do you not think the manuscript was a doozy? Why not?

      I did not mean to attack the paper, but I did mean to offer a critical opinion. Which of my ‘attacks’ did you think were incorrect or wrong-headed?

  3. What strikes me as interesting is that you actually make rather ignorant statements about the publication (as if you haven’t, in fact, read nor understood it) which renders this entire blog nonsense. Are you aware of how ignorant the statements you make sound? Are you not conscious of how petty you appear? Are you confronted with your own inadequate research which makes it all the more important to try to tear down the research of others? To what end? Are you proud of yourself? Does it make you feel like a better person to try to ransack the achievements of others?

    • What did you find to be ignorant? Can you offer specific criticisms? I am happy to be shown to be wrong.

      I was not trying to…ransack…their achievements, but I do have my own thoughts and opinions! I do not have to automatically agree with *everything* they claim in the paper, do I? I’d love to hear your specific thoughts on the paper

    • BenH, obviosly you are offended, but to me his statements didn’t seem that ignorant. Maybe a little superficial, but hey its a blog, not a paper.

      • I don’t know what a troll is (apart from the creature that lives under bridges) but I did explain myself below. This blog’s sole purpose was to undermine the research of this group. If you’re going to do that, at least do it from an objective, scientific perspective. That’s the bottom-line of scientific respect. I don’t understand the younger generation who think that because you’re able to post things and garner some sort of following that you natively are entitled to consequence-free-public-judgement of someone else’s work…. But the consequences will/do catch up. Also, it’s likely you will continue to get ‘trolls’ (have I understood what a troll in this context is, correctly?)

        Let me ask you a question – if this was your work, if it was your scientific life that was being undermined by a thoughtless ‘superficial blog post written in a coffee shop’ by someone who is not an expert in your field but claims to be one and to understand your work. How would it make you feel? Would you feel you had been respectfully critiqued? Or would you feel that someone had taken 20 minutes of his life to undermine decades of yours?

  4. Everybody is a critic, and some of them are doozies (so I lose my PhD now?)

    My main criticisms are A) knowing what the neurons do in the in vitro slice tells us frack all about what’s going on in vivo. Basic example – hippocampal neurons in vivo have an input resistance ~10 fold lower than in vitro. Why? Because there’s so damn much going on via active inputs and neuromodulators and B) at p14 these rats are very immature. Whole cell types will be missing and firing behavior is likely completely different than adult, though some cells are likely mature at p14, others are far from it..

  5. Just read the comments section to your doozy of a blogpost … some mad passionate Markram fanfolks out there. Agree with your assessment of the Markram project, despite your h-index, and admire your professional responses to the ad hominem unprofessional attacks by Eva and BenH. h-index based attacks?!?!?!

    • I just think that if you are going to take a critical stance on someone else’s life work that you should at the very least, respect them enough to thoroughly read and understand their work. Perhaps you’re still too young to know it but one day, many years from now, when you’ve spent a lifetime trying to find a truth or make a difference, you’ll then know how important it is that people respect your work at the very least, they don’t have to agree with it, nor take it up and criticism is critical to moving forward and improving. But at least respect it. As scientists, that’s the very least we can do for each other. I say that your blog was premature and ignorant because you muddled even the most basic concept of the paper which clearly shows that you did a very quick read through or didn’t actually read it – how they classified their neurons. M type – morphological type (various types of neuron morphological they found). E type – electrical type (various electrical patterns they found which happen to appear in many different types of neurons). Me type – the morphological type that matches a specific electrical type…. The rest of your statements also tend to show a basic/shallow understanding of what they actually did… This is someone’s life work, don’t undermine it unless you have a proper and thorough knowledge. However, when I read your blog, I was surprised and shocked that someone could make light of this work without taking into account what it must mean for these people that put their blood into it and I was outraged. Yet, I should not have been so harsh. I apologize for that.

      • I am certainly not going to disagree that this was a superficial post. It was just some thoughts I wrote out in a coffeeshop.

        But I certainly understand the paper and the ideas behind it. Markram plan has been around for, what, a decade now? This certainly is not the first time I have seen them, or discussed them. In fact, there are many, many criticisms that I did not lay out – questions that I have seen Markram personally field, so he is quite obviously aware of them – that do not get resolved in the paper.

        In the specifics of the paper, like I said, I got the paper well before it came out. I discussed it with people. I read the paper from beginning to end, read the companion papers, and understand the figures. That’s a lot! One might call it a doozy. If you want me to *thoroughly* go through the paper on this blog, I can? But covering the detail was not what I was trying to do here.

        I respect that people have put a lot of time and effort into this. That this is their life. And they did a lot of good stuff here! But conceptually, this is a very controversial plan of attack.

        (fwiw, using terms like ‘m-type’ instead of just ‘morphology’ smacks too much of obscurantism to me. How did I muddle that? You’ll note that the two figures I included in this post were showing m-types and e-types. But I spent too much time reading and arguing about philosophy as an undergrad to think positively of making up terms when you don’t need to do so.)

      • BenH, it’s absurd on its face to suggest that it’s beyond the pale for a blogger (or anyone else) to criticize the output of the Blue Brain Project unless one has read everything that came before it. By that standard, virtually no one is qualified to comment on anyone else’s work. I can count the number of times on zero fingers that I’ve bothered to read everything a given author has produced before reading one of their papers—and that includes papers that editors have sent me with the explicit request that I review them as a supposed expert in an area. A scientific publication is meant to be a largely self-contained work that assumes a grounding in the methods and key results of a field, but does not critically depend on prior work. If this particular 100+ page paper cannot be understood by a researcher with a grounding in computational neuroscience who has carefully read all of the text and the supplement, then frankly, the problem lies with the paper, not with the reader.

        Of course, I’m mostly speculating about what it is you don’t like about Adam’s critique, because you haven’t bothered to explain what your specific objection is to any of the claims in the post. Is it not true that one might be able to find almost anything in the model given the flexibility of the implementation? Well, then perhaps you can provide clear examples from the paper of phenomena that the model was unable to reproduce. Do you disagree that there are a large number of undocumented free parameters involved in generating results that match the observed dynamics of rat neurons? Perhaps you could provide an accounting of how the parameters of the model were arrived at, or provide an indication of how many experiments were conducted and not reported in the manuscript. Don’t like the assertion that the divergence between in silico and in vivo conductances is problematic? Then maybe you can explain why it’s not a principled problem when such data diverge from one’s expectations. Think it’s unfair to characterize the Blue Brain project as “a social conundrum for the field”? I invite you to re-assess the many comments here and elsewhere to the same effect, not to mention the various political problems that have arisen at various stages of the process. Frankly, from where I’m sitting, there seems to be nothing deeply objectionable about this post, and if anything, its main limitation is that it largely echoes concerns people have expressed elsewhere.

        As to your question about how Adam would feel if it was his work that was “being undermined”: I can’t speak for Adam, but I suppose my own answer would be, “why does it matter?” Since when is science about protecting anyone’s feelings? I mean, let’s suppose for the sake of argument that the Blue Brain Project is a colossal waste of taxpayer funds (I’m not saying this is true, but let’s play out the hypothetical). Do you suppose in that case that there would be any conceivable way to make that argument in a public setting without hurting Henry Markram’s feelings, no matter how many of his papers one had previously read? Should we avoid critiquing a billion Euro project just because one or more of the hundreds of people involved with the project in some capacity appear to have a rather thin skin? Speaking for myself, I’d like to think that if someone wrote a blog post that I thought “undermined” my work in an unfair way, my reaction would be to (a) get upset about it for a little while, then (b) get over it and go write a carefully reasoned response explaining why I thought the arguments in question were fallacious. I’d like to believe that I wouldn’t resort to appeals to authority or name-calling. But even if I’m wrong about the latter point, it still wouldn’t have any bearing on the matter. The fact that your feelings appear to be hurt by this post is quite irrelevant to scientific progress unless you can also point out what is substantively wrong about it. If you can’t, you’re just wasting everyone else’s time. Virtually every scientific manuscript ever produced is a serious investment of someone’s time, and any number of manuscripts represent the culmination of many years or decades of work. One can respect the effort and intention of another researcher and simultaneously view that researcher’s output as flawed and deserving of criticism. If it were any other way, science would grind to a halt. Markram and his colleagues are not sufficiently special that only researchers who’ve read their entire oeuvres are qualified to comment on the Blue Brain Project.

  6. Pingback: Spike activity 09-10-2015 « Mind Hacks

  7. Pingback: Spike activity 09-10-2015 | Chatag

  8. In my view, the biggest shortcoming of this kind of work is that it operates with only one source of the constraints: the neural architecture. Understanding how a complex system works requires an understanding born from many kinds of constraints that cooperatively reduce the space of permissible theories to a manageable number. By focussing exclusively on one type of constraint, the search for good theory is going to be much less efficient, because there are many (<– dramatic understatement) neurally plausible architectures that exhibit implausible behavior.

  9. “Il est dangereux d’avoir raison dans des choses où des hommes accrédités ont tort.” (translation: it is dangerous to be right in matters where established men are wrong.)
    — Voltaire [in “Catalogue pour la plupart des écrivains français qui ont paru dans Le Siècle de Louis XIV, pour servir à l’histoire littéraire de ce temps,” Le Siècle de Louis XIV (1752)]

    Thank you @neuroecology for the quality of your replies to some of these comments.

  10. Pingback: Recording thousands of cells like it’s nobody’s business | neuroecology

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s