Refine
Year of publication
- 2024 (2) (remove)
Document Type
- Article (2) (remove)
Language
- English (2)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2)
Keywords
- Adult neurogenesis (1)
- Calpain (1)
- Deep learning (1)
- Digital pathology (1)
- Glioma (1)
- MEIS homeodomain protein (1)
- Stem cell (1)
- Subventricular zone (1)
- Transcription factor (1)
- Transfer learning (1)
Institute
- Medizin (2)
We provide in this paper a comprehensive comparison of various transfer learning strategies and deep learning architectures for computer-aided classification of adult-type diffuse gliomas. We evaluate the generalizability of out-of-domain ImageNet representations for a target domain of histopathological images, and study the impact of in-domain adaptation using self-supervised and multi-task learning approaches for pretraining the models using the medium-to-large scale datasets of histopathological images. A semi-supervised learning approach is furthermore proposed, where the fine-tuned models are utilized to predict the labels of unannotated regions of the whole slide images (WSI). The models are subsequently retrained using the ground-truth labels and weak labels determined in the previous step, providing superior performance in comparison to standard in-domain transfer learning with balanced accuracy of 96.91% and F1-score 97.07%, and minimizing the pathologist's efforts for annotation. Finally, we provide a visualization tool working at WSI level which generates heatmaps that highlight tumor areas; thus, providing insights to pathologists concerning the most informative parts of the WSI.
Tight control over transcription factor activity is necessary for a sensible balance between cellular proliferation and differentiation in the embryo and during tissue homeostasis by adult stem cells, but mechanistic details have remained incomplete. The homeodomain transcription factor MEIS2 is an important regulator of neurogenesis in the ventricular–subventricular zone (V-SVZ) adult stem cell niche in mice. We here identify MEIS2 as direct target of the intracellular protease calpain-2 (composed of the catalytic subunit CAPN2 and the regulatory subunit CAPNS1). Phosphorylation at conserved serine and/or threonine residues, or dimerization with PBX1, reduced the sensitivity of MEIS2 towards cleavage by calpain-2. In the adult V-SVZ, calpain-2 activity is high in stem and progenitor cells, but rapidly declines during neuronal differentiation, which is accompanied by increased stability of MEIS2 full-length protein. In accordance with this, blocking calpain-2 activity in stem and progenitor cells, or overexpression of a cleavage-insensitive form of MEIS2, increased the production of neurons, whereas overexpression of a catalytically active CAPN2 reduced it. Collectively, our results support a key role for calpain-2 in controlling the output of adult V-SVZ neural stem and progenitor cells through cleavage of the neuronal fate determinant MEIS2.