Visualizing the brain networks with neuTube

Ting Zhao, developer of the neuTube software, talks about his tool and its benefits for users.

brain-image

A software for neuron reconstruction and visualization

Why did you develop neuTube?        

The goal of developing the neuTube software is to help neuroscientists reconstruct digital neuronal morphology efficiently from fluorescence microscope images.

Digital reconstruction, or tracing, of neuron morphologies from light microscope images is an important step of mapping brain circuits. Despite numerous algorithms proposed to trace neurons automatically, proof-editing is often necessary to obtain acceptable neuron structures, and it could be a bottleneck to neuron reconstruction. Quite a few software tools have been developed to solve this problem. The efficiency of editing, however, is still limited by editing capabilities in those tools. Especially, the lack of software that integrates comprehensive functions of visually editing a neuron structure has kept users from correcting various types of error with intuitive operations. To address these issues, we have designed a comprehensive editing framework based on the SWC model and implemented it in the neuTube software.

Feature highlights of neuTube software tool

What about the quality of neuTube: availability, usability and flexibility?

Compiled versions of neuTube for different operating systems, including Microsoft Windows, Mac OSX and Linux, are available for free.  We also provide complete software documentation and video tutorials. The source code of neuTube can be downloaded from GitHub. The authors are very responsive to regarded questions. NeuTube can be used to reconstruct neurons in 3D fluorescence image data from scratch, or edit standard SWC files produced by any other software.

Could you give 5 reasons to choose neuTube for neuron network analysis and visualization?

Powerful visualization: neuTube provides high-quality visualization of images and morphological data in both 2D and 3D. Especially, we optimized our 3D visualization engine to render millions of neuron nodes in real time.

Robust tracing algorithms: neuTube has two semi-automatic tracing methods: local tracing and path searching. The local tracing algorithm, which has won us a prize in the DIADEM challenge, enables tracing a whole neuronal branch with a single mouse click. Path searching allows the user to extend branches by specifying target points. These two complementary approaches are combined to minimize mouse clicks in tracing neurons with complex morphologies.

User-friendly interface: Editing a neuron structure in neuTube is simple. The user just need to select a set of nodes and then apply an operation though the context menu or a shortcut key. Advanced functions such as selecting nodes based on tree topology or automatically connect nodes nearby are also provided to speed up editing. All the editing operations are tracked by the software to support unlimited undo/redo functionality, which, absent in most free neuron tracing software, is essential for tolerating user mistakes.

Editing in both 2D and 3D: Most editing functions are available in both 2D and 3D views. The 2D view is good for examining details and the 3D view makes most operations intuitive.  All changes will be synchronized between the 2D and 3D views.

Tested in practice: We have used neuTube to reconstruct many beautiful neurons in several of our papers (for example in Structured Synaptic Connectivity between Hippocampal Regions).  It has also been used by quite a few labs around the world.

neuTube-figure

Mapping brain connectivity with neuTube 1.0 from a New Design for Efficient Neuron Reconstruction Software Based on the SWC Format.

About the author:

Ting Zhao is the co-author with Linqing Feng, of neuTube 1.0: a New Design for Efficient Neuron Reconstruction Software Based on the SWC Format.

He is a senior software engineer of the FlyEM Project at Janelia Research Campus, Ashburn, Virginia, US, and was recently awarded in the DIADEM competition. With many years of experience in image analysis and software development, Ting Zhao is interested in developing computational methods, algorithms and software tools for revealing how the brain works.

Based of the recent paper:

(Feng et al., 2015) neuTube 1.0: A New Design for Efficient Neuron Reconstruction Software Based on The SWC Format. eNeuro.