A Poisson generalized linear model application to disentangle the effects of various parameters on neurophysiological discharges
| Autor: | Stefano Diomedi, Francesco Edoardo Vaccari, Patrizia Fattori, Matteo Filippini, Claudio Galletti |
|---|---|
| Přispěvatelé: | Vaccari Francesco Edoardo, Diomedi Stefano, Filippini Matteo, Galletti Claudio, Fattori Patrizia |
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Generalized linear model
Science (General) Computer science Bioinformatics Action Potentials Neurophysiology Single Cell Kinematics Poisson distribution General Biochemistry Genetics and Molecular Biology symbols.namesake Q1-390 Model Organisms Position (vector) Protocol Animals Poisson Distribution Protocol (object-oriented programming) Bioinformatic General Immunology and Microbiology Model Organism General Neuroscience Fingerprint (computing) Correction Brain Signal Processing Computer-Assisted Metric (mathematics) symbols Linear Models Macaca Algorithm Neuroscience |
| Zdroj: | STAR Protocols, Vol 2, Iss 2, Pp 100413-(2021) STAR Protocols, Vol 2, Iss 4, Pp 100930-(2021) STAR Protocols STAR Protoc |
| ISSN: | 2666-1667 |
| Popis: | Summary The protocol provides an extensive guide to apply the generalized linear model framework to neurophysiological recordings. This flexible technique can be adapted to test and quantify the contributions of many different parameters (e.g., kinematics, target position, choice, reward) on neural activity. To weight the influence of each parameter, we developed an intuitive metric (“w-value”) that can be used to build a “functional fingerprint” characteristic for each neuron. We also provide suggestions to extract complementary useful information from the method. For complete details on the use and execution of this protocol, please refer to Diomedi et al. (2020). Graphical abstract Highlights • GLM applied to neuronal discharges reveals parameters that modulate their activity • Applying regularizers helps to discard minority parameters and reduces noise • Each neuron can be characterized by the weight assigned to each parameter tested • Results are well suited for population analyses (i.e., clustering, correlation, etc) The protocol provides an extensive guide to apply the generalized linear model framework to neurophysiological recordings. This flexible technique can be adapted to test and quantify the contributions of many different parameters (e.g., kinematics, target position, choice, reward) on neural activity. To weight the influence of each parameter, we developed an intuitive metric (“w-value”) that can be used to build a “functional fingerprint” characteristic for each neuron. We also provide suggestions to extract complementary useful information from the method. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|