In [18]:

HTML("""
<h1 class="title">Course on Computational Neuroscience</h1>
<h2>PhD program, Marseille</h2>
<h1> December 7th & 8th, 2015</h1>
<img src="figures/troislogos.png" width=61%/>
""")

Out[18]:

Course on Computational Neuroscience

PhD program, Marseille

December 7th & 8th, 2015

Alessandro Torcini, INS, INMED & CPT
Demian Battaglia, INS
Laurent Perrinet, INT
Wahiba Taouali, INMED

Acknowledgements:¶

PhD program: Anna Montagnini, Frédéric Chavane, Nadia Pittet, INT, Marseille

ⓦ https://laurentperrinet.github.io/sciblog/files/2015-12-08_cours_neurocomp/PerrinetEtAl12neurocomp-intro.slides.html

(CONF) Hello, I am Laurent Perrinet, https://laurentperrinet.github.io/sciblog/files/2015-12-08_cours_neurocomp/PerrinetEtAl12neurocomp-intro.slides.html
(SHOW AUTHORS) on behalf of the people that will give the course, I
(ACKNO) From the head on, I wish to thanks people who collaborated on this endeavour, ....
(SHOW TITLE) CompNeuro is interested in the link between the neural activity and its function. in particular, for vision, I am researching the relation between the functional organization (anatomy and activity) of low-level visual areas and the structures that appear to be in natural scenes, that is of the images that hit the retina and which are relevant to us.
(OBJECTIVE) The course aims at introducing students with the major tools that will be necessary during their thesis to model or analyze their neuroscientific results. While it will start by a short, generic introduction, we will then explore different systems at different scales. On the first day, we will study the different possible regimes in which a single neuron can behave, while progressively introducing the theory of dynamical systems to understand these more globally. Then, during the second day, we will introduce methods to analyze neuroscientific data in general, such as Bayesian methods and information theory. This will be implemented by simple practical examples.

In [19]:

#Image('http://static1.squarespace.com/static/55249fa2e4b004d7b9c37cf8/t/553d629be4b0810e9e82b01f/1430086332477/Neural.jpg', height=height, embed=True)
Image('figures/Neural.jpg', height=height, embed=True)
# http://www.mth.kcl.ac.uk/research/cns/brain.gif 
# http://www.nbb.cornell.edu/Courses/image/Bionb3300%20Linster.jpg

Out[19]:

Computational Neuroscience¶

A branch of neuroscience which uses mathematical and computer models to understand how the brain works, that is, the principles of computation and representation and their neural implementation.
Aims:
- what? description: unify data in a single framework.
- how? understand underlying mechanisms.
- why? understand principles underlying functions (optimality for eg) ⇢ make predictions / guide experiments. better data analysis.

In [20]:

#Image('https://www0.maths.ox.ac.uk/files/imce/u896/homer_0.jpg', height=height, embed=True)
Image('figures/homer_0.jpg', height=height, embed=True)

Out[20]:

Some history¶

In [21]:

HTML("""
<img src="figures/levels.png" height=1200 />
What is computational neuroscience?  Patricia S. Churchland, Christof Koch, Terrence J. Sejnowski. in Computational Neuroscience pp.46-55. Edited by Eric L. Schwartz. 1993. MIT Press
""")
#Image('figures/levels.png', height=height, embed=True)

Out[21]:

What is computational neuroscience? Patricia S. Churchland, Christof Koch, Terrence J. Sejnowski. in Computational Neuroscience pp.46-55. Edited by Eric L. Schwartz. 1993. MIT Press

In [22]:

#Image('http://www.gatsby.ucl.ac.uk/~Dayan/book/cover.gif', height=height, embed=True)
#Image('http://ecx.images-amazon.com/images/I/4127RGJX8NL._SY344_BO1,204,203,200_.jpg', height=height, embed=True)
Image('figures/4127RGJX8NL._SY344_BO1,204,203,200_.jpg', height=height, embed=True)

Out[22]:

In [23]:

#Image('http://www.paris-neuroscience.fr/sites/paris-neuroscience.fr/files/cns2013_poster.jpg', height=height, embed=True)
Image('figures/cns2013_poster.jpg', height=height, embed=True)

Out[23]:

The term "computational neuroscience" was introduced by Eric L. Schwartz, who organized a conference, held in 1985 in Carmel, California, at the request of the Systems Development Foundation to provide a summary of the current status of a field which until that point was referred to by a variety of names, such as neural modeling, brain theory and neural networks. The proceedings of this definitional meeting were published in 1990 as the book Computational Neuroscience.[2] The first open international meeting focused on Computational Neuroscience was organized by James M. Bower and John Miller in San Francisco, California in 1989 and has continued each year since as the annual CNS meeting [3] The first graduate educational program in computational neuroscience was organized as the Computational and Neural Systems Ph.D. program at the California Institute of Technology in 1985.

Major topics in Computational Neuroscience¶

A branch of neuroscience which uses mathematical and computer models to understand how the brain works, that is, the principles of computation and representation and their neural implementation.
Aims:
- what? description: unify data in a single framework.
- how? understand underlying mechanisms.
- why? understand principles underlying functions (optimality for eg) ⇢ make predictions / guide experiments. better data analysis.

In [24]:

Image('figures/computation-02-00047-ag_bigger.jpg', height=height, embed=True)

Out[24]:

In [25]:

#Image('https://dipanjanr.files.wordpress.com/2015/04/figure11.jpg', height=height, embed=True)
Image('figures/figure11.jpg', height=height, embed=True)
#Image('https://dipanjanr.files.wordpress.com/2015/04/rp_fig2-copy-page-001-1.jpg', height=height, embed=True)

Out[25]:

Dipanjan Roy, Rodrigo Sigala, Michael Breakspear, Anthony Randal McIntosh, Viktor K. Jirsa, Gustavo Deco, Petra Ritter. Revealing how local and global plasticity shapes the brain’s dynamical landscape. Brain Connectivity Journal 01 oct 2014 dx.doi.org/10.1089/brain.2014.0252.

RP_Fig2 copy-page-001 (1)The human brain is a complex system capable of producing nonstationary spatio temporal signals. Mathematical descriptions based on neural mass models describing population firing rate and time-dependent analysis of regional time series is capable of making predictions about systems dynamics. This further establishes a direct bridge between biologically inspired theory, simulations and experimental design. This informed prediction from theory based on biological constraints serves as an important tool for designing novel sensory stimuli to probe brain dynamics at multiple spatial and temporal scales of organization. In our lab we examine structural and functional brain networks using data from non-invasive neuroimaging techniques (fMRI, MEG, MRI, DTI, DSI). Our goal is to determine fundamental organizational principles of both underlying anatomy and specificity of functional dynamics. Our results collectively point to principles of optimal topology of networks that supports certain function of modules, spatial and temporal scaling of network organization, and network adaptability in response to increasing cognitive demands or in the context of learning. We are also interested in the recently emerging field of computational neuropsychiatry where complementary evidence accumulates from neuropsychiatric disease, specifically schizophrenia, Parkinsonian disease that exhibits disruption of normal connectivity patterns, the prevalence of wiring inefficiency, disruption of neurochemical balance and as a consequence impact directly whole brain network dynamics.

In [26]:

#Image('https://upload.wikimedia.org/wikipedia/en/f/ff/Monkey_using_a_robotic_arm.jpg', height=height, embed=True)
#Monkey operating a robotic arm with brain–computer interfacing (Schwartz lab, University of Pittsburgh) 
#Image('http://www.nature.com/nrn/journal/v15/n5/images_article/nrn3724-f1.jpg', height=height, embed=True)
Image('figures/nrn3724-f1.jpg', height=height, embed=True)

Out[26]:

The program¶

See https://laurentperrinet.github.io/sciblog/files/2015-12-08_cours_neurocomp/PerrinetEtAl12neurocomp-intro.slides.html

Day 1 : 2015-12-07 : Computational Neuroscience¶

09:15-09:45 = An introduction to the field of Computational Neuroscience (LP)
09:45-12:30 = Introduction to modeling single neurons: the Hodgkin-Huxley model (AT)

14:00-15:30 = An introduction to neural masses: modeling assemblies of neurons up to capturing resting state dynamics in a mean-field model - presentation of the Virtual Brain software (DB)
16:00-17:00 = A concrete application of modeling neural masses: the epileptor, or "From mathematical entomology to the virtual epileptic patient" (DB)

Day 2 : 2015-12-08 : Information theory / statistics of decoding¶

09:00-10:30 = An overview on "What is encoding?" "What is decoding?": formalization of the notion of information in neural activity (DB)
11:00-12:30 = Live information! From sharing information to transferring information (and a glimpse into the zoo of higher-order friends) (DB)
14:00-17:30 = Decoding of feature selectivity in neural activity: Concrete applications in visual data (WT & LP).