In this paper, we address the scene segmentation task by capturing rich contextual dependencies based on the self-attention mechanism. Unlike previous works that capture contexts by multi-scale feature fusion, we propose a Dual Attention Network (DANet) to adaptively integrate local features with their global dependencies. Specifically, we append two types of attention modules on top of dilated FCN, which model the semantic interdependencies in spatial and channel dimensions respectively. The position attention module selectively aggregates the feature at each position by a weighted sum of the features at all positions. Similar features would be related to each other regardless of their distances. Meanwhile, the channel attention module selectively emphasizes interdependent channel maps by integrating associated features among all channel maps. We sum the outputs of the two attention modules to further improve feature representation which contributes to more precise segmentation results. We achieve new state-of-theart segmentation performance on three challenging scene segmentation datasets, i.e., Cityscapes, PASCAL Context and COCO Stuff dataset. In particular, a Mean IoU score of 81.5% on Cityscapes test set is achieved without using coarse data. 1 .

SUMMARY IDH1 and IDH2 mutations occur frequently in gliomas and acute myeloid leukemia, leading to simultaneous loss and gain of activities in the production of α-ketoglutarate (α-KG) and 2-hydroxyglutarate (2-HG), respectively. Here we demonstrate that 2-HG is a competitive inhibitor of multiple α-KG-dependent dioxygenases, including histone demethylases and the TET family of 5-methlycytosine (5mC) hydroxylases. 2-HG occupies the same space as α-KG does in the active site of histone demethylases. Ectopic expression of tumor-derived IDH1 and IDH2 mutants inhibits histone demethylation and 5mC hydroxylation. In glioma, IDH1 mutations are associated with increased histone methylation and decreased 5-hydroxylmethylcytosine (5hmC). Hence, tumor-derived IDH1 and IDH2 mutations reduce α-KG and accumulate an α-KG antagonist, 2-HG, leading to genome-wide histone and DNA methylation alterations.

No abstract