Whole-brain mapping of neuroplasticity in different experimental paradigms in mice - a computational perspective

• Autorzy: Bednarek S. , Jermakow N. , Stefaniuk M. , Pawlowska M. , Wojcik D.K. , Majka P.
• Języki publikacji: EN

Abstrakty

EN

INTRODUCTION: Imaging of entire brains at cellular resolution, enabled by light-sheet fluorescence microscopy (LSFM) and optical tissue clearing, offers insights into neural activity at a high magnification while preserving the brain‑wide context. AIM(S): We propose a set of open-source computational tools that address three fundamental challenges associated with the analysis of LSFM images of entire rodent brains, namely: management of voluminous imaging data, alignment to a reference atlas, and object detection and localization. METHOD(S): The data for each brain, such as multichannel acquisitions and spatial information, are compressed and stored in an HDF5-based container as a pyramid of resolutions to facilitate and standardize data access and manipulation. Unlike most other alignment approaches, our pipeline is not only guided by standard similarity metrics such as mutual information, but also utilizes Deep Convolutional Neural Networks to generate label maps corresponding to specific brain structures such as main white matter tracts or dentate gyrus. This step significantly increases the accuracy and robustness of the registration procedure. The c‑Fos‑positive nuclei are identified and quantified with the help of another neural network trained on synthetic data, generated to simulate the original nuclei which eliminated the laborious process of manual image annotation. The software was applied to investigate c-Fos-mediated neuroplasticity in iDISCO-cleared brains in experimental paradigms of appetitive and aversive learning and alcohol addiction. RESULTS: Voxel-wise statistical analysis revealed brain areas involved in the neuroplasticity of alcohol addiction and appetitive or aversive learning in mice. CONCLUSIONS: We demonstrate the ability of our software to combine efficient data management, accurate atlas alignment, and object detection to facilitate LSFM analyses. FINANCIAL SUPPORT: ERA‑NET NEURON/17/2017 grant from NCBR, G2631 grant from NCN.

• Wydawca: -
• Czasopismo: Acta Neurobiologiae Experimentalis
• Rocznik: 2019
• Tom: 79
• Numer: Suppl.1
• Opis fizyczny: p.LVIII-LIX

Twórcy

autor

Bednarek S.

• Laboratory of Neuroinformatics, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland

autor

Jermakow N.

• Laboratory of Neuroinformatics, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland

autor

Stefaniuk M.

• Laboratory of Neurobiology, BRAINCITY, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland

autor

Pawlowska M.

• Laboratory of Neurobiology, BRAINCITY, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland

autor

Wojcik D.K.

• Laboratory of Neuroinformatics, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland

autor

Majka P.

• Laboratory of Neuroinformatics, Nencki Institute of Experimental Biology, Polish Academy of Sciences, Warsaw, Poland