Experimental quality control induces changes in Allen mouse brain connectomes

Area of Science:

• Neuroscience
• Connectomics
• Computational Biology

Background:

• The Allen Mouse Brain Connectivity Atlas (AMBCA) is a key resource for studying mouse brain structural connectivity.
• Existing connectomes derived from AMBCA data may contain inaccuracies due to experimental limitations.

Purpose of the Study:

• To perform rigorous quality control (QC) on AMBCA experiments.
• To rebuild mouse brain connectomes using a curated dataset.
• To assess the impact of QC on connectome architecture.

Main Methods:

• Combined automated and manual quality control (QC) applied to 437 AMBCA experiments.
• Identification and exclusion of 56 experiments with injection or projection failures.
• Reconstruction of regionalized voxel and homogeneous connectomes using 381 validated experiments.

Main Results:

• Identified ~13% of experiments as having significant quality issues.
• Rebuilt connectomes show altered connectivity patterns, including losses and gains between brain regions.
• Analysis revealed subtle changes in whole-brain organization, such as rich club and community structure.

Conclusions:

• Rigorous QC is essential for accurate representation of mouse brain structural connectivity.
• The rebuilt connectomes, particularly the regionalized voxel model, offer improved accuracy.
• Corrected connectomes provide a more reliable foundation for future connectomic studies.