Connectomics — Connecting the Dots of our Brains!

9 min readOct 26, 2022

One of the first things I had to learn as a kid was how to write in cursive. Also no, not because it was required of us in school, but rather because my indian parents considered cursive writing to be among the most important skills a child could have. I was given books to use that had pages full of tiny, dotted letters, and I was instructed to trace each one for an hour in order to practise writing in cursive.

This continued for many months, possibly even a full year or two. I eventually learned cursive writing, but I continued to trace those tiny dotted lines. I progressed to increasingly challenging tasks, such as drawing cartoon characters like Mickey Mouse or Cinderella by connecting the dots. And as time passed and I kept working on connecting all those little dots, I discovered that this ability could be applied on a 10x scale.

It turns out adults have been using this skill to find chronic neurological and psychiatric disorders in the brain and understand the intricacy of neurological functions. And they’ve been doing this through the study of connectomics.

The study of connectomics is a fairly new thing, and was pioneered by something called the “Human Connectome Project”. The Human Connectome Project, or HCP for short, intends to construct a complete connectome, which is a map of the 100 billion neurons in the human brain. The connectome may also include the synaptome (synapses) and epigenome (epigenetic factors), depending on the neuroscientist you ask.

Biology 101

Back to the Basics

Before you’re able to understand the way connectomics works, you need to fully understand the brain, and the neurons in your brain. Our nervous systems sense what is happening both outside of us and inside of us; they then decide how we should respond and change the condition of our internal organs (such as changing our heart rates). They also enable us to think about and remember what has happened in our lives (our memories) It uses neurons, which are a relatively sophisticated network, to do this.

The brain contains an estimated 86 billion neurons, each of which is linked to 1,000 more neurons to form an immensely intricate communication network. The smallest functional component of the nervous system is a neuron (nerve cell). Basically Neurons work like this; If the inputs are receptive and threshold is reached (-55 mV), a nerve impulse will travel along the length of the axon. The dendrites receive inputs from the axon terminals of other neurons. Neurotransmitters will then be released into synapses, (the 20 nm extracellular gap between two neurons), as a result of this.

This diagram just showcases the structure of one out of the many types of different neurons in the brain. To put things in perspective, just the cerebellum itself contains the following different kinds of neurons:

Cerebellum Lugaro cell
Cerebellum Golgi cell
Cerebellum basket cell
Cerebellum nucleus reciprocal projections neuron
Cerebellum stellate cell
Cerebellum unipolar brush cell

Reminder - these are just a few to name.

In other words, the Human Connectome Project is examining the brain in a way that has never been done before. This is essential because, to put it mildly, the tools we currently use to to research and learn the brain can be pretty inaccurate and aren’t always the best option.

How does the HCP work?

Connectomes can be divided into two categories: macroscale and microscale. Macroscale connectomics is the process of mapping out huge fibre tracts and functional grey matter regions in the brain in terms of blood flow (functional) and water diffusivity using structural and functional MRI data (structural). Microscale connectomics involves using histology and microscopy to map the whole connectome of small organisms. Specifically, all of the connections in their central nervous system.

Macroscale connectomics inside the human brain are evaluated using magnetic resonance imaging. White matter tracts are mapped using dMRI image series, and blood flow correlations between related grey matter regions are evaluated using fMRI data.

Macroscole Connectomics

Macroscale connectomes are normally collected using imaging technologies like diffusion magnetic resonance imaging (dMRI) and functional magnetic resonance imaging (fMRI). White matter between the cortex and subcortex can be imaged using dMRI datasets that can cover the entire brain. As a metric of neural activation, cerebral blood flow is measured in fMRI datasets.

One of the benefits of MRI is it offers in vivo (in vivo meaning tests and experiments that are performed in or on a whole living organism) information about connectivity between different brain areas. Our understanding of multiple brain networks, including the visual, brainstem, and language networks, among others, has improved thanks to macroscale connectomics.

Microscale Connectomics

Microscale connectomes, on the other hand, concentrate on a considerably smaller region of the nervous system with much higher resolution. These datasets, which provide single-synapse resolution of whole local circuits, are frequently gathered using electron microscope imaging. The complete nervous system of C. elegans, an entire fly brain, and most recently a millimetre cube from both mouse and human cortex are some of the greatest accomplishments in EM connectomics.

Mesoscale Connectome

Now you might be wondering, we’ve covered microscale and macroscale connectomes but what’s a mesoscale connectome?

This type of connectome is a lot less common to see in the connectomics field, but nevertheless is another amazing type of connectome!
A mesoscale connectome is a general area of connection between groups of similar neurons (individual neurons cannot be distinguished).

The Automatic Tape-Collection Mechanism

This odd looking device is known as the Automatic Tape-Collection Mechanism and it automatically records brain tissue on thick plastic film. Every wire is on its own section of the brain thanks to the ultramicrotome cutting arm, which moves the brain against a diamond knife and cuts each section by 30 nm. The sections are picked up by the conveyor belt after floating on water. The device can cut 10,300 brain sections daily, which is more than 100 TB of data! This seems like a lot but in reality is only about a grain of salt in size.

After the sections are picked up by a conveyor belt, they’re then cut individually and placed onto silicon wafers.

If you’ve ever seen a movie in the making, you probably understand how tedious the process of cutting all the scenes together is!

When imaged under an electron microscope, each part of the brain can be compared to a scene from a movie. However, if we apply the same analogy, we would need 33,333 scenes to create a cubic millimetre of a movie — which is a lot.

One 3D cubic millimeter of the brain made from 33,333 2D images of sliced brain

This data was used by researchers to follow a section of a neuron. This is what was left after everything around it was taken away:

But here’s the problem: this was neither feasible nor efficient at all, requiring wayyyyyy too much work for such a tiny amount of data. On all 33,333 images, the neuron had to be manually coloured in by the researchers. This is what hand segmentation, also known as “Color in the Dots,” or hand colouring all the parts of a single frame, looks like:

But luckily, computers are helping us out with this annoying and time-consuming process. Artificial intelligence is now being used to fully automate the segmentation process!

1,000 cubic microns of everything around a single dendrite of a pyramidal neuron

There are a grand total 675 synapses, 530 axons, and 90 dendrites in the image above. It shows exceptionally non-random connectivity (P =.00002), indicating that some axons favour innervating particular dendrites. To improve the effectiveness of this careful and thorough procedure, microscopes like a 61 beam scanning electron microscope are being developed right now.

Connection Matrices

So through all these different processes, we acquire a lot of data. But once we have this what do we do with it?

We make things called Brain Networks! Simply put, brain networks are accumulations of nodes (neuronal components) and edges (interconnections). Depending on the way the neuroscientists define a “node,” they are obtained from measurements of the structural or functional relationships between pairs of neurons or brain regions.
The connection matrix, which shows the topology of the network, is shown in the diagram below. The edges are the lines, while the nodes are the circles.

Important things to note

If a node has few connections, it is referred to as being of “low degree.” Having multiple connections makes a node “high degree”
Two modules can be used to further separate a network
A high degree node may be a provincial hub or a connector hub, maintaining connections between various modules within a single region within the brain.

We can use specialised statistical tools from graph theory with the matrix thats created to our advantage! Segregation, integration, and influence are the three broad categories that make up measures of brain connectivity.

We’ll explore segregation and integration in more detail.

The clustering coefficient of the network or its natural tendency to form distinct modules are two ways to express the amount that nodes accumulate into separate clusters in measures of segregation.
Key integration metrics relate to communication efficiency and path length, which are important aspects of how nodes can exchange information.
Measures of integration quantify the ease with which communication occurs along network paths.

Basically, the presence of a “small world” network is indicated by the combination of high clustering (segregation) and short path length (integration), as seen above.

These small world networks can be analysed in patients with various different disorders to truly understand them neurologically: We can develop more effective treatments by properly understanding the underlying causes of these complex disorders (and who wouldn’t want that?).

Why does any of this matter anyway?

The mapping of the brain’s neuronal connectivity is one potential application of connectomics in the field of neuroscience. This might shed light on the connections between various parts of the brain and offer a potential cure for diseases like autism or Alzheimer’s. Making precise models of the brain is a further potential application of connectomics.

This could help researchers better understand the brain, improve education, or develop more human-like artificial intelligence. In personalised medicine, where treatments are specifically tailored to a person based on their individual brain connectivity, connectomics may also be used. This could be used to treat illnesses like epilepsy or schizophrenia that are brought on by abnormal brain connectivity.

In essence, the field of “connectomics” is tremendously important. We could develop more effective and efficient treatments and solutions for neurodegenerative illnesses like Alzheimer’s and psychiatric disorders with a deeper understanding of the brain. We might even be able to give illnesses or even memories a physical appearance with the use of connectomes!