Chondrosarcoma (CHS) is a heterogeneous collection of malignant bone tumours and is the second most common primary malignancy of bone after osteosarcoma. Recent work has identified frequent, recurrent mutations in IDH1/2 in nearly half of central CHS. However, there has been little systematic genomic analysis of this tumour type and thus the contribution of other genes is unclear. Here we report comprehensive genomic analyses of 49 cases of CHS. We identified hypermutability of the major cartilage collagen COL2A1 with insertions, deletions and rearrangements identified in 37% of cases. The patterns of mutation were consistent with selection for variants likely to impair normal collagen biosynthesis. In addition we identified mutations in IDH1/2 (59%), TP53 (20%), the RB1 pathway (33%) and hedgehog signaling (18%).

BackgroundEukaryotic mitogen-activated protein kinase (MAPK/MPK) signaling cascades transduce and amplify environmental signals via three types of reversibly phosphorylated kinases to activate defense gene expression. Canola (oilseed rape, Brassica napus) is a major crop in temperate regions. Identification and characterization of MAPK and MAPK kinases (MAPKK/MKK) of canola will help to elucidate their role in responses to abiotic and biotic stresses.ResultsWe describe the identification and analysis of seven MKK (BnaMKK) and 12 MPK (BnaMPK) members from canola. Sequence alignments and phylogenetic analyses of the predicted amino acid sequences of BnaMKKs and BnaMPKs classified them into four different groups. We also examined the subcellular localization of four and two members of BnaMKK and BnaMPK gene families, respectively, using green fluorescent protein (GFP) and, found GFP signals in both nuclei and cytoplasm. Furthermore, we identified several interesting interaction pairs through yeast two-hybrid (Y2H) analysis of interactions between BnaMKKs and BnaMPKs, as well as BnaMPK and BnaWRKYs. We defined contiguous signaling modules including BnaMKK9-BnaMPK1/2-BnaWRKY53, BnaMKK2/4/5-BnaMPK3/6-BnaWRKY20/26 and BnaMKK9-BnaMPK5/9/19/20. Of these, several interactions had not been previously described in any species. Selected interactions were validated in vivo by a bimolecular fluorescence complementation (BiFC) assay. Transcriptional responses of a subset of canola MKK and MPK genes to stimuli including fungal pathogens, hormones and abiotic stress treatments were analyzed through real-time RT-PCR and we identified a few of BnaMKKs and BnaMPKs responding to salicylic acid (SA), oxalic acid (OA), Sclerotinia sclerotiorum or other stress conditions. Comparisons of expression patterns of putative orthologs in canola and Arabidopsis showed that transcript expression patterns were generally conserved, with some differences suggestive of sub-functionalization.ConclusionsWe identified seven MKK and 12 MPK genes from canola and examined their phylogenetic relationships, transcript expression patterns, subcellular localization, and protein-protein interactions. Not all expression patterns and interactions were conserved between canola and Arabidopsis, highlighting the limitations of drawing inferences about crops from model species. The data presented here provide the first systematic description of MKK-MPK-WRKY signaling modules in canola and will further improve our understanding of defense responses in general and provide a basis for future crop improvement.

Hypoxia in solid tumors is thought to be an important factor in resistance to therapy, but the extreme microscopic heterogeneity of the partial pressures of oxygen (pO 2) between the capillaries makes it difficult to characterize the scope of this phenomenon without invasive sampling of oxygen distributions throughout the tissue. Here we develop a noninvasive method to track spatial oxygen distributions in tumors during fractionated radiotherapy, using oxygen-dependent quenching of phosphorescence, oxygen probe Oxyphor PtG4 and the radiotherapy-induced Cherenkov light to excite and image the phosphorescence lifetimes within the tissue. Mice bearing MDA-MB-231 breast cancer and FaDu head neck cancer xenografts show different pO 2 responses during each of the 5 fractions (5 Gy per fraction), delivered from a clinical linear accelerator. This study demonstrates subsurface in vivo mapping of tumor pO 2 distributions with submillimeter spatial resolution, thus providing a methodology to track response of tumors to fractionated radiotherapy.

The automatic diagnosis of various retinal diseases from fundus images is important to support clinical decisionmaking. However, developing such automatic solutions is challenging due to the requirement of a large amount of humanannotated data. Recently, unsupervised/self-supervised feature learning techniques receive a lot of attention, as they do not need massive annotations. Most of the current self-supervised methods are analyzed with single imaging modality and there is no method currently utilize multi-modal images for better results. Considering that the diagnostics of various vitreoretinal diseases can greatly benefit from another imaging modality, e.g., FFA, this paper presents a novel self-supervised feature learning method by effectively exploiting multi-modal data for retinal disease diagnosis. To achieve this, we first synthesize the corresponding FFA modality and then formulate a patient feature-based softmax embedding objective. Our objective learns both modality-invariant features and patient-similarity features. Through this mechanism, the neural network captures the semantically shared information across different modalities and the apparent visual similarity between patients. We evaluate our method on two public benchmark datasets for retinal disease diagnosis. The experimental results demonstrate that our method clearly outperforms other self-supervised feature learning methods and is comparable to the supervised baseline. Our code is available at GitHub 1. Index Terms-Retinal disease diagnosis, self-supervised learning, multi-modal data I. INTRODUCTION C OLOR fundus photography has been widely used in clinical practice to evaluate various conventional ophthalmic diseases, e.g., age-related macular degeneration (AMD) [1], pathologic myopia (PM) [2], and diabetic retinopathy [3, 4]. Recently, deep learning has shown very good performance on a variety of automatic ophthalmic disease detection problems from fundus images [5-7], and these techniques can help ophthalmologists in decision making. The success is attributed to the learned representative features from fundus images, which requires a large amount of training data with massive human annotations. However, it is tedious and expensive to annotate the fundus images, since experts are needed to provide reliable labels. Hence, in this paper, our goal is to learn the representative features from data itself, without any human

No abstract