Predicting Neural Connections Using Calcium Channel Imaging
Neurons and their Networks
Neuron
Figure 1 - Neuron Structure (wikimedia.org)
Neuron is an electrically excitable cell and we can also it known as the basic unit of a nervous system. Neurons can processes and transmit information through electrical and chemical signals. Neurons are major components of the brain. The main function of a neuron is receiving input from other neurons at the synapses (signals between neurons occur via specialized connections) and if required conditions are satisfied, produce an output through the axon.
Neurons can be classified by the direction that they send information,
1. Sensory neurons
​These neurons can send information from skin, eyes, nose like sensory receptors toward the central nervous system.
2. Motor neurons
These neurons can send information away from the central nervous system to muscles or glands.
3. Inter neurons
These neurons can send information between sensory neurons and motor neurons.
Also we can classify neurons by their shape,
Multipolar neuron
Bipolar neuron
Pseudo-unipolar neuron
Unipolar neurons
Figure 2 - Basic Neuron Types (google.com)
Then we get know about neural network,
Neural network
Figure 3 - Conntections between neurons (chalearn.org)
Neural network is an information processing model that is inspired by the biological nervous systems such as brain, process information. Simply we can say that neural network is an interconnection of neurons. By using neural network we can show connections between neurons and their neighbor neurons. In neural network connections have directions.
Identify the connections between neurons by imaging the calcium influx
In the previous topic, I discussed about what is a neuron, types of neurons and the connections between that neurons. Now I’m going to discuss about how to identify and measure the connections between neurons by imaging the calcium influx into neurons using fluorescent molecules. First we get to know about the calcium imaging.
Calcium Imaging
Calcium imaging is a scientific technique that designed to show the calcium (Ca2+) status of a cell, tissue or medium. In the calcium imaging technique fluorescent molecules that can respond to the binding of Ca2+ ions by changing their fluorescence properties. There are two types of calcium indicators,
Chemical indicators
Genetically encoded calcium indicators
These calcium imaging can be used for optically probe intracellular calcium in living animals and also these calcium imaging allowed to study about the neural activities in millions of neurons. Calcium ions generate versatile intracellular signals that control key functions in all types of neurons. Below figure about Neuronal Calcium Signaling.
Figure 1 - Neuronal Calcium Signaling
Now let’s know about neural activities using calcium fluorescence
Neural activities
The sense of neuronal activity is outside the brain ... brain activity can be understood, that is, it makes sense only insofar as it is a representation of the state of the body, the external world or a potential behavior. Just as it would be useless to analyze a book investigating the chemical composition of paper and ink, a reductionist analysis of brain activity, disassembling and analyzing its nucleic acids, enzymes, receptors and ion channels, fails to explain what brain activity makes.
By watching following video you can get an abstract idea about neural activity.
[youtube https://www.youtube.com/watch?v=OJh7FszRsgo&w=854&h=480]
Figure 1 - Video of neural activity visualized with calcium fluorescence in a neuron culture.
The Best Method To Predicting Connections Between Neurons Using Calcium Channel Imagine
Time Series
Time series is a series of data points listed or graphed in successive order of time.
In this method we can choose a variable changes with the. By analyzing time series can examine how the changes associated with considering compare to in variables the time period. we are predicting connections between neurons we use this method by following way.
Figure 1 - Time series plot of quakes
First we compute the connectivity matrix. The entries of these particular matrix map to excitatory connections according to time series signals which exist between pairs of neurons inside the brain. It consists of two steps, construction the connectivity matrix and global thresholding. Finding the possibility of connectivity can find the excitatory connections between each pair of neurons.
When we are computing the connectivity matrix we need another method of Transfer Entropy.
Transfer Entropy
We can take transfer entropy method as the suitable method to predicting the connections between neurons. The definition of the transfer entropy is a non statistic measuring the amount of time asymmetric transfer of information between two random processes. By following equation we can find the transfer entropy.
The equation always depend on the times series.
Where,
Xn is the value of the time series X at the time n.
Yn is the value of the time series Y at the time n.
Xn(k) = Xn , Xn-1 , Xn-k-1 and Yn(k)= Yn , Yn-1 , Yn-k-1
are the k previous events.
By using the above equation we can measures the difference between the distribution of the next value of the sequence X given its own history and the distribution of the next values of X given its own history and the history of Y.
We can see from above equation if the X does not depend of Y the result will be 0. That means Transfer entropy equals to zero.
If the X depend of Y then the transfer entropy is grater than zero.
By transfer entropy we can that X is depend or not of Y but we can’t identify the connections between X and Y. For that purpose we need modified transfer entropy method.
Modified Transfer Entropy Method
where,
Xn is the differential fluorescence level of neuron X at the time n.
Yn is the differential fluorescence level of neuron Y at the time n.
gn is the average differential fluorescence level of the network at time t.
g’ is a threshold
By this method using the following equation we can find the g’ value.
If the g’ is greater than zero then there is a connection between neuron Y to neuron X. This is the most efficient and better way to predicting the connections between neurons.
Finally using calcium channel imagine and the modified transfer method we can predict neuron connections.