Characterizing Synaptic Protein Clusters

Who are we?

The Synaptic Protein Zoo project is managed by members of the Flavie Lavoie-Cardinal lab at the CERVO Brain Research Centre in Quebec City, Canada. Our work is multidisciplinary: it combines neurobiology, optics and machine learning to study molecular phenomena within neurons. Using cutting edge microscopy techniques, we observe biological phenomena at a nanoscopic scale, which we can subsequently analyze using state of the art artificial intelligence approaches.

What are synaptic proteins?

Synapses are where neurons communicate to allow us to think, move and feel. At synapses, neurons send signals to each other using molecules called neurotransmitters. The signal is transmitted from a presynaptic neuron to a postsynaptic neuron. The complex mechanisms of neurotransmitter release and reception are controlled by molecules called synaptic proteins. Synaptic proteins are an interesting subject to study as their organisation within neurons is an indicator of the strength of synapses and is modified when the brain learns. Furthermore, brain diseases can affect this organisation. Therefore, studying synaptic proteins may help us to learn more about the molecular mechanisms of brain diseases and allow for better treatments to be made. Amyotrophic lateral sclerosis, better known as ALS, and Parkinson's disease are examples of neurodegenerative diseases that affect the health of synapses and that can be studied by looking at synaptic protein organization.

a) Synapses are the connections formed between a presynaptic neuron's axon at the bouton and a postsynaptic neuron's dendrite at the spine. Presynaptic proteins are found in the bouton (magenta) and postsynaptic proteins are found in the dendritic spine (green). b) Cartoon representation of a synapse in different orientations with the pre- and postsynaptic proteins highlighted. c) Note that typical microscopy images of synaptic proteins are in two dimensions, which means that synapses may appear side by side (top), diagonally (middle), or one on top of the other (bottom).

How do we look at synaptic proteins?

Synapses are very small: the distance between both neurons at a synapse is around 30 nanometers, which is over 3000 times thinner than a human hair. At this scale, specialized equipment is required to be able to distinguish two small things that are very close together. We use a special microscope called STED which allows us to see where specific molecules are located at a very small scale. The following image is an example of synaptic proteins imaged using the STED microscope:

On this image, every colored blob corresponds to synaptic proteins concentrated within a synapse. We call these blobs clusters. The shape, size and position of these clusters are what we are interested in for this study. However, it is difficult to characterize such a large number of clusters, which is where you come in!

Why do we need your help?

Synaptic protein images are difficult to analyze due to the variability in the sizes and shapes of the clusters and due to their large quantity. One of our images can contain between 500 and 2000 clusters of varying shapes and sizes. While there are methods available to analyze this type of data, we would prefer a new approach that requires less tinkering with parameters and that analyzes a wider variety of elements. We want to use machine learning to be able to automatically analyze different characteristics of synaptic protein clusters. To do so, we will require large amounts of annotated data to train a good model. However, our images are very difficult to annotate manually in large quantities by only a few people. This is why we want to use crowdsourcing to create a large dataset of annotated synaptic protein images with which to train machine learning models that will be able to do this task automatically.

How can you help?

Simply click on the "Classify" tab to begin labeling synaptic protein clusters! Please view the tutorials to better understand what to do and what to look out for. If you have any questions, take a look at the FAQ tab, the Field Guide, or use the Talk section to let us know. Thank you for taking the time to help us in our studies!