Limits...
A statistical method of identifying interactions in neuron-glia systems based on functional multicell Ca2+ imaging.

Nakae K, Ikegaya Y, Ishikawa T, Oba S, Urakubo H, Koyama M, Ishii S - PLoS Comput. Biol. (2014)

Bottom Line: The interactions in our interest included functional connectivity and response functions.We evaluated the cross-validated likelihood of GLMs that resulted from the addition or removal of connections to confirm the existence of specific neuron-to-glia or glia-to-neuron connections.We only accepted addition or removal when the modification improved the cross-validated likelihood.

View Article: PubMed Central - PubMed

Affiliation: Integrated Systems Biology Laboratory, Graduate School of Informatics, Kyoto University, Sakyo-ku, Kyoto, Japan.

ABSTRACT
Crosstalk between neurons and glia may constitute a significant part of information processing in the brain. We present a novel method of statistically identifying interactions in a neuron-glia network. We attempted to identify neuron-glia interactions from neuronal and glial activities via maximum-a-posteriori (MAP)-based parameter estimation by developing a generalized linear model (GLM) of a neuron-glia network. The interactions in our interest included functional connectivity and response functions. We evaluated the cross-validated likelihood of GLMs that resulted from the addition or removal of connections to confirm the existence of specific neuron-to-glia or glia-to-neuron connections. We only accepted addition or removal when the modification improved the cross-validated likelihood. We applied the method to a high-throughput, multicellular in vitro Ca2+ imaging dataset obtained from the CA3 region of a rat hippocampus, and then evaluated the reliability of connectivity estimates using a statistical test based on a surrogate method. Our findings based on the estimated connectivity were in good agreement with currently available physiological knowledge, suggesting our method can elucidate undiscovered functions of neuron-glia systems.

Show MeSH

Related in: MedlinePlus

Neuron–glia network estimated from Ca2+ imaging data.(A) Connectivity matrix of the neuron–glia network estimated with our method. Each column and each row of the matrix correspond to “sender” (i.e., from) neuron/glia and “receiver” (i.e., to) neuron/glia. Indices of 48 neurons and indices of six glial cells are segmented by white lines on the matrix. Each matrix entry denotes the root mean square of the corresponding response function; the root mean square is normalized within the entry values of  and  individually. This is because the magnitude of the response functions was considerably different across  and . For example, the element (1, 49) indicates the magnitude of the response function from neuron 1 to glial cell 1 (). (B) The proportions (as percentages) of the response functions,  and , which took positive values are depicted in the top left, top right, bottom left, and the bottom right panels, respectively. The self-feedback connections of neurons, represented by , were all inhibitory, which would demonstrate the refractoriness of neurons. On the other hand, the self-feedback connections of glial cells, represented by , were all excitatory. This could be due to the timescale of the glial activities, which are much slower than the sampling frequency.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4230777&req=5

pcbi-1003949-g003: Neuron–glia network estimated from Ca2+ imaging data.(A) Connectivity matrix of the neuron–glia network estimated with our method. Each column and each row of the matrix correspond to “sender” (i.e., from) neuron/glia and “receiver” (i.e., to) neuron/glia. Indices of 48 neurons and indices of six glial cells are segmented by white lines on the matrix. Each matrix entry denotes the root mean square of the corresponding response function; the root mean square is normalized within the entry values of and individually. This is because the magnitude of the response functions was considerably different across and . For example, the element (1, 49) indicates the magnitude of the response function from neuron 1 to glial cell 1 (). (B) The proportions (as percentages) of the response functions, and , which took positive values are depicted in the top left, top right, bottom left, and the bottom right panels, respectively. The self-feedback connections of neurons, represented by , were all inhibitory, which would demonstrate the refractoriness of neurons. On the other hand, the self-feedback connections of glial cells, represented by , were all excitatory. This could be due to the timescale of the glial activities, which are much slower than the sampling frequency.

Mentions: Fig. 3A shows the identified connectivity matrix of the neuron–glia network. Here, we assumed that the functional connections between neurons and glia were directional because the neuron-to-glia and the glia-to-neuron connections are believed to depend on different biophysical processes [23]. There are small numbers of connections with substantially larger values than the other connections at the top left of the matrix, i.e., inter-neuronal connections. This observation is consistent with existing physiological studies, which report that the strength of inter-neuronal connections in the hippocampus obeys a log-normal distribution [42] We can also see some strong glia-to-neuron connections at the top right.


A statistical method of identifying interactions in neuron-glia systems based on functional multicell Ca2+ imaging.

Nakae K, Ikegaya Y, Ishikawa T, Oba S, Urakubo H, Koyama M, Ishii S - PLoS Comput. Biol. (2014)

Neuron–glia network estimated from Ca2+ imaging data.(A) Connectivity matrix of the neuron–glia network estimated with our method. Each column and each row of the matrix correspond to “sender” (i.e., from) neuron/glia and “receiver” (i.e., to) neuron/glia. Indices of 48 neurons and indices of six glial cells are segmented by white lines on the matrix. Each matrix entry denotes the root mean square of the corresponding response function; the root mean square is normalized within the entry values of  and  individually. This is because the magnitude of the response functions was considerably different across  and . For example, the element (1, 49) indicates the magnitude of the response function from neuron 1 to glial cell 1 (). (B) The proportions (as percentages) of the response functions,  and , which took positive values are depicted in the top left, top right, bottom left, and the bottom right panels, respectively. The self-feedback connections of neurons, represented by , were all inhibitory, which would demonstrate the refractoriness of neurons. On the other hand, the self-feedback connections of glial cells, represented by , were all excitatory. This could be due to the timescale of the glial activities, which are much slower than the sampling frequency.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4230777&req=5

pcbi-1003949-g003: Neuron–glia network estimated from Ca2+ imaging data.(A) Connectivity matrix of the neuron–glia network estimated with our method. Each column and each row of the matrix correspond to “sender” (i.e., from) neuron/glia and “receiver” (i.e., to) neuron/glia. Indices of 48 neurons and indices of six glial cells are segmented by white lines on the matrix. Each matrix entry denotes the root mean square of the corresponding response function; the root mean square is normalized within the entry values of and individually. This is because the magnitude of the response functions was considerably different across and . For example, the element (1, 49) indicates the magnitude of the response function from neuron 1 to glial cell 1 (). (B) The proportions (as percentages) of the response functions, and , which took positive values are depicted in the top left, top right, bottom left, and the bottom right panels, respectively. The self-feedback connections of neurons, represented by , were all inhibitory, which would demonstrate the refractoriness of neurons. On the other hand, the self-feedback connections of glial cells, represented by , were all excitatory. This could be due to the timescale of the glial activities, which are much slower than the sampling frequency.
Mentions: Fig. 3A shows the identified connectivity matrix of the neuron–glia network. Here, we assumed that the functional connections between neurons and glia were directional because the neuron-to-glia and the glia-to-neuron connections are believed to depend on different biophysical processes [23]. There are small numbers of connections with substantially larger values than the other connections at the top left of the matrix, i.e., inter-neuronal connections. This observation is consistent with existing physiological studies, which report that the strength of inter-neuronal connections in the hippocampus obeys a log-normal distribution [42] We can also see some strong glia-to-neuron connections at the top right.

Bottom Line: The interactions in our interest included functional connectivity and response functions.We evaluated the cross-validated likelihood of GLMs that resulted from the addition or removal of connections to confirm the existence of specific neuron-to-glia or glia-to-neuron connections.We only accepted addition or removal when the modification improved the cross-validated likelihood.

View Article: PubMed Central - PubMed

Affiliation: Integrated Systems Biology Laboratory, Graduate School of Informatics, Kyoto University, Sakyo-ku, Kyoto, Japan.

ABSTRACT
Crosstalk between neurons and glia may constitute a significant part of information processing in the brain. We present a novel method of statistically identifying interactions in a neuron-glia network. We attempted to identify neuron-glia interactions from neuronal and glial activities via maximum-a-posteriori (MAP)-based parameter estimation by developing a generalized linear model (GLM) of a neuron-glia network. The interactions in our interest included functional connectivity and response functions. We evaluated the cross-validated likelihood of GLMs that resulted from the addition or removal of connections to confirm the existence of specific neuron-to-glia or glia-to-neuron connections. We only accepted addition or removal when the modification improved the cross-validated likelihood. We applied the method to a high-throughput, multicellular in vitro Ca2+ imaging dataset obtained from the CA3 region of a rat hippocampus, and then evaluated the reliability of connectivity estimates using a statistical test based on a surrogate method. Our findings based on the estimated connectivity were in good agreement with currently available physiological knowledge, suggesting our method can elucidate undiscovered functions of neuron-glia systems.

Show MeSH
Related in: MedlinePlus