Back

Navigating the Molecular Maze of Alzheimer's Disease: Developing A Human Hippocampal Atlas to Bridge Discovery and Clinical Practice

Wang, P.; Wang, L.; Tao, Q.; Guo, Z.; Luo, T.; Xu, Z.; He, Y.; Yu, J.; Liu, Y.; Wu, Z.; Xu, B.; Jin, B.; Yang, Y.; Cheng, M.; Jiang, Y.; Tian, C.; Zheng, H.; Fan, Z.; Jiang, P.; Gao, Y.; Wu, J.; Wang, S.; Sun, B.; Wei, Y.; Fang, Z.; Lei, J.; Luo, B.; Wen, H.; Peng, G.; Tang, Y.; Su, X.; Pan, C.; Zhu, K.; Shen, Y.; Liu, S.; Bao, A.; Yao, J.; Wang, J.; Xu, X.; Li, X.; Liu, L.; Duan, S.; Han, L.; Zhang, J.

2023-12-07 neurology
10.1101/2023.12.06.23299593 medRxiv
Show abstract

Alzheimers disease (AD), the primary cause of dementia worldwide, persists as an enigma, with the specific molecular and cellular mechanisms driving cognitive decline still eluding understanding. To aid in unraveling this enigma, we generated a spatial transcriptomic atlas of the human hippocampus with single-cell resolution, leveraging integration of spatial mapping with single-cell transcriptomic profiles from six individuals with or without AD. Utilizing this complex and multi-dimensional data, the atlas pinpoints AD-associated transcriptomic changes with spatial specificity, such as a pronounced elevation of inflammatory responses and energy metabolism in fimbria and CA4, respectively. AD also caused alterations of the cellular composition in hippocampal subregions, even around the amyloid-beta (A{beta}) plaques. Finally, our investigation uncovered pathways intricately linked to synthesis and secretion of extracellular vesicles (EVs). Within these altered pathways, we identified several corresponding proteins, including CCK and PMP2, that are readily detectable in human blood, holding promise for aiding AD diagnosis in clinical settings.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.