Connectomics

• Responsibles: Henry Kennedy (DRE INSERM), Colette Dehay (DRCE CNRS)
• Pierre Misery -Engineer (IE) 25%
• Camille Lamy -Engineer (IE) 20%

A connectome is a comprehensive map of the connections of the nervous system. The Connectomics Platform is more specifically concerned with the investigation of the cortical connectome at the meso-scale. That is to say the comprehensive set of connections linking cortical areas in two model organisms, the macaque and the mouse. The fact that the connectome deals with comprehensive maps, means that there is a need to develop high-throughput analysis of connectivity. The platform is geared to analyzing the inputs to defined cortical areas that are determined using high-fidelity, high-sensitivity retrograde tracers .

The cortical connectome can be viewed as a graph and the analysis of the cortical graph exploits definitions and algorithms used in graph theory. The density of the cortical graph is very high (macaque 67%, mouse 97%). However, the range of weights of connections span 5 to 6 orders of magnitude, meaning that binary, connected/not-connected measurements are of little value. The determination of the weighted cortical graph requires accurate counts of labeled neurons in order to generate reliable and consistent connectivity profiles of individual areas.

The Connectomics Platform consists of three interrelated facilities: (i) a customized surgical suite for injection of tracers; (ii) a histology lab for preparing material for microscopic examination; (iii) microscope facilities.

Mouse and macaque databases that have been constructed by this platform can be obtain here : http://www.core-nets.org.

Surgical suite

Fully equipped for investigating the anesthetized macaque in a stereotaxic apparatus (respirator, CO2 monitor, blood pressure and CO2 monitor, heating blanket). Tracer injections are made using BrainSight® technology from Rogue Research Inc, a brain navigator tool that exploits MRI scans acquired in the 3-T Siemens MRI scanner in Primage, imaging platform annexed to SBRI.

Histology lab

Tissue is prepared in a fully equipped histology lab with chemicals and glassware. The histology lab has linear hooded mounting desks for mounting sections and freezers and refrigerators for storing antibodies, and appropriate reagents for immunohistochemistry. The histology lab is equipped with a large microtome for cutting NHP brains. The Microm HM 440 E Microtome is an ergonomically-designed sliding microtome. It employs electronic control systems and high-precision mechanics to achieve optimal section quality and reproducibility of paraffin and frozen specimens. It can accommodate specimen sizes up to 80 x 60 mm, and can cut sections ranging from 1 to 100 microns.

Microscope facilities

Our automated imagery stations exploit ExploraNova® software for neuronal cartography in order to explore projection pathways. There are 4 fluorescent microscopy stations running on the Mercator program. Analysis consists in the recording of the XY coordinates of very large numbers of neurons using a digitalized imaging system equiped with automatic cell counting. 2D cartography of neuronal labeling is used to reconstruct 3D (Map3D) maps to obtain accurate data on individual neuron pathways length, important for modeling networks of the cerebral cortex. One of the stations exploits the Morphostrider® software from ExploraNova®, which allows automatic 2D reconstructions of large and high-defined images with multiple wave-lengths.

In addition to charting neurons using ExploraNova® technology we also use slide scanning technology. Zeiss AxioscanZ1 uses multiple channels to create virtual digital copies of sections with up to 4 different tracers that are detected at different wave-lengths and subsequently analyzed using Mercator.

1 / Microscope Leica Dialux 22 with fluorescence

Objectives : x2,5 x10 x25 x50 x100
Filter : D
Camera Sony DXC950-P
Mercator expert software

2 / Microscope Leica DMRE with fluorescence

Objectives : x1,6 x5 x10 x20 x25 x40
Camera Jenoptik ProgRes CF
Filters : XF102-2, N2-1, XF100-2, D, XF13-2, XF103
Mercator expert software.

3 / Microscope Leica DM5500 B with fluorescence.

Objectives : x2,5 x10 x20 x40.
Filters : A, D, Y5, I3 and N2-1.
Camera Jenoptik ProgRes CF.
Mercator expert, Map3D and  Morphostrider Software.

4 / Microscope Leica DM6000 with fluorescence

Objectives : x2,5 x10 x20 (dry) x20 (imm) x63 x100
Filters : TX2, D, DY, FB, I3, XF32, XF08
Camera Jenoptik ProgRes CF.
Mercator expert software.

Data Analysis

The analysis of the structural connectivity network is done using the free software environment for statistical computing and graphics R (https://cran.r-project.org/). In particular, the majority of network properties are investigated using network analysis tools available in the R package igraph (http://igraph.org/r/). For data visualization we mainly use the packages lattice (https://cran.r-project.org/web/packages/lattice/index.html) and latticeExtra (https://cran.r-project.org/web/packages/latticeExtra/index.html). We use the package stats (https://stat.ethz.ch/R-manual/R-devel/library/stats/html/00Index.html) for additional statistical calculations. We are currently developing an R toolbox for analyzing complex structural networks (directed, undirected, weighted and unweighted), in particular to compute commonly used measures of centrality, segregation and influence, at the local and global levels.