Weiqi Zhu

In this paper, a graph-based nonlocal total variation method is proposed for unsupervised classification of hyperspectral images. The variational problem is solved by the primal-dual hybrid gradient algorithm. By squaring the labeling function and using a stable simplex clustering routine, an unsupervised clustering method with random initialization can be implemented. The effectiveness of this proposed algorithm is illustrated on both synthetic and real-world HSI, and numerical results show that the proposed algorithm outperforms other standard unsupervised clustering methods such as spherical K-means, nonnegative matrix factorization, and the graph-based Merriman-Bence-Osher scheme.

Lineage progression and diversification is regulated by the coordinated action of unique sets of transcription factors. Oligodendrocytes and astrocytes comprise the glial sub-lineages in the CNS, and the manner in which their associated regulatory factors orchestrate lineage diversification during development and disease remains an open question. Sox10 and NFIA are key transcriptional regulators of gliogenesis associated with OL and AS. We found that NFIA inhibited Sox10 induction of OL differentiation through direct association and antagonism of its function. Conversely, we found that Sox10 antagonized NFIA function and suppressed AS differentiation in mouse and chick systems. Using this developmental paradigm as a model for glioma, we found that this relationship similarly regulated the generation of glioma subtypes. Our results describe the antagonistic relationship between Sox10 and NFIA that regulates the balance of OL and AS fate during development and demonstrate for the first time, to the best of our knowledge, that the transcriptional processes governing glial sub-lineage diversification oversee the generation of glioma subtypes.