AxoFluidic

Estrela Neto^1,2, Meriem Lamghari^1,3 & Paulo Aguiar⁴

¹ INEB – Instituto de Engenharia Biomédica, Rua do Campo Alegre, 823 4150-180 Porto, Portugal

² FMUP - Faculdade de Medicina da Universidade do Porto, Porto, Portugal

³ ICBAS - Instituto de Ciências Biomédicas Abel Salazar, Universidade do Porto, Porto, Portugal

⁴ Centro de Matemática da Universidade do Porto, Porto, Portugal

AxoFluidic Release

The AxoFluidic software is available here:

- MATLAB files (for users with MATLAB): AxoFluidic_MATLAB.zip

- Standalone (MATLAB not required): AxoFluidic_64bits_R2012b.exe

To run the standalone it is necessary to have the MATLAB™ Component Runtime (MCR) installed. This is a free package from Mathworks which can be downloaded from here (MCR).

The authors make NO WARRANTY or representation, either express or implied, with respect to this software, its quality, accuracy, merchantability, or fitness for a particular purpose. This software is provided "AS IS", and you, its user, assume the entire risk as to its quality and accuracy.

For comments and bugs reports please use pauloaguiar at fc.up.pt

AxoFluidic related reference

Important: If you use the AxoFluidic software please reference the journal published in Integrative Biology journal:

Neto E, Alves CJ, Sousa DM, Alencastre IS, Lourenço AH, Leitão L, Ryu HR, Jeon NI, Fernandes R, Aguiar P, Almeida RD, Lamghari M. Sensory neurons and osteoblasts: close partners in a microfluidic platform. DOI: 10.1039/b000000x

About

In the field of neuroscience, analyses often rely upon neuro-anatomical tracing of axonal processes, requiring a significant amount of user-dependent image processing and may result in fatigue-related user bias. The AxoFluidic software was developed with the objective of quantify the axonal outgrowth within compartmentalized microfluidic platforms, where three distinct domains can be easily identified: somal side – microgrooves – axonal side (Fig1).

The AxoFluidic quantifies the axonal density spatial dependence using a moving column travelling across the longitudinal axis of the image. The column has a predefined width, which balances the desired spatial resolution of the measurement with the spatial scale of heterogeneities.

The algorithm works on a binary mask of the axonal plexus. The binary masks are obtained using composite morphological operations in order to convert the axonal plexus into a wireframe with a single pixel diameter. Under these conditions, the total amount of non-zero pixels in a column provides a reasonable estimate of the total length of axonal fibers. The spatial profile produced provides a complementary assessment in conditions where large sprouting may occur. The mean amplitude of the measurement in the somal side is used for normalizing the spatial profile amplitude. This normalization provides the means to compare the transition from somal side to axonal side between different experiments.

In the specific operation of the AxoFluidic software the user is asked to upload 8-bit image and define the limits of the somal and axonal borders. An “Increase Contrast” bottom is available to enhance the contrast and help define the microgrooves limits.

Images corresponding to the neuronal marker are first preprocessed to remove background gradients through the use of morphological structuring elements “Morph Analysis”.

The algorithm to quantify the axonal density spatial dependence uses a moving column travelling across the longitudinal axis of the image. The column has a predefined width (adjusted by the user) which balances the desired spatial resolution of the measurement with the spatial scale of heterogeneities.

After profile calculation (pressing “Calculate Profile”) the exponential fit is given by the spatial dependence decay function f(x) = A × exp(–x/λ), where A is an estimate of the fraction of axons that can effectively cross the microgrooves, x is the spatial variable and λ returns the scale of spatial decay (in µm). Larger λ values indicate more extensive axonal growth. The axonal density decreases by 1/e (e := exp(1); roughly 1/3) after λ distance.

The blue profile is the normalized density of axons present in the somal side (0-400 µm, in this example). In green is the axonal density profile in the microgrooves region (400-700 µm, in this example) and in red in the axonal density profile in the axonal side (700-1500 µm, in this example). The horizontal blue line indicates the normalized (average) density value in the axonal side) and the red line shows the exponential fit to the density profile of the axonal growth.

Installation

The AxoFluidic software was implemented using MATLAB™ and is distributed both as source code and as a MS Windows™ 64 bit compiled executable. No support is available at this moment for other operating systems.

To run AxoFluidic from MATLAB call the script AxoFluidic.m availiable in AxoFluidic_MATLAB.zip.

To run the AxoFluidic standalone (.exe version), install first the MATLAB™ Component Runtime (MCR) available here (MCR), and then double-click the downloaded executable AxoFluidic_64bits_R2012b.exe.

For comments and bugs reports please use pauloaguiar at fc.up.pt

Project grant

This software results from a collaboration with the Neuro-osteogenesis Team led by Meriem Lamghari, from INEB's NEWTherapies Group and Paulo Aguiar from Centro de Matemática da Universidade do Porto. The work by Meriem Lamghari was co-funded by the Fundação para a Ciência e aTecnologia (FCT), FEDER and COMPETE-Program POFC/QREN.