Ming Zhang

Optimization of immuno-precipitation-based neo-antigen enrichment.

MANA-SRM sensitivity improvement through the D-Red approach.

Supplementary Figure and Table Legends

Visual localization is a core component of technologies enabling autonomous driving, augmented reality, etc. Image retrieval is a crucial step commonly used in state-of-the-art visual localization approaches. This paper presents a text-based image retrieval and filtering algorithm specifically designed for shopping mall scenarios. Compared to repetitive architectural styles, e.g. walls, floor, columns, which are common in shopping malls, logos of the shops are more informative and distinctive. The characteristics of text regions are fully utilized: (1) features extracted by a CNN network are used to filter out texts with less information, (2) image patches centered on text areas are used for global feature extraction, (3) the size and contact of text areas are used for the ranking of retrieved images when multiple texts are detected in one query image. Our approach can be incorporated into most existing retrieval network generating a global descriptor for the input image. Intensive experiments are carried out on ground-truth image databases, verifying the effectiveness of our approach.

Embryonic stem cells (ESCs) are characterized by their ability to self-renew and their potential to differentiate into any cell type. Therefore, identification of novel molecular markers to verify the pluripotent status of mouse ESCs (mESCs) is of great significance. Kinase D interacting substrate of 220 kDa (Kidins220)/ankyrin repeat-rich membrane spanning (ARMS) plays a crucial role in the integration of growth factor receptor pathways during embryonic development. However, the role of Kidins220/ARMS in ESCs is still unknown. To elucidate the effects of Kidins220/ARMS on ESCs, we performed a knockdown of the Kidins220/ARMS gene by RNA interference. To our surprise, downregulation of Kidins220/ARMS did not alter the pluripotent state of mESCs. In contrast, it was essential for the proliferation and survival of ESCs. Furthermore, downregulation of the ARMS gene limited the migration of embryoid body cells derived from mESCs. This study indicates novel roles of Kidins220/ARMS in ESCs, which may represent valuable targets for future clinical applications of ESCs.

This paper presents a system for performance-driven avatar animation by estimating facial pose and expression parameters from single image. In this system, a 3D shape prediction model is trained based on local binary feature (LBF) algorithm, which use random forest to extract image features and learns a linear regression model mapping these features to 3D shape. With the help of this model, the 3D positions of facial vertexes can be estimated from a web camera image. The facial pose and expression parameters can be calculated by solving an optimization problem that fitting a set of blend shapes models to these 3D vertexes. Experiments show that our system can estimate accurate facial parameters from single image and generate similar looking avatar animation.

Mesenchymal stromal cells (MSCs) are one of the most intensively studied stem cell types with application aims. However, the molecular characterisation and the relationship between the molecular characterisation and functional properties of MSCs are largely unknown. In this study, we purified the surface proteins from rat bone marrow MSCs (rBMMSCs) and characterised their surface proteome by LC-MS/MS. Moreover, we comparatively analysed the data from this study with the surface proteomics data of mouse and human embryonic stem (ES) cells and human mesenchymal stromal cells (hMSCs). The data showed that, in contrast to ES cells and human mesenchymal stromal cells, rBMMSCs possessed a surface proteomics pattern biased to neural and neural-endocrine lineages, indicating a neural/neural crest bias, and suggested a neural differentiation tendency of these cells. The different surface proteomics pattern between rBMMSCs and hMSCs also suggested that MSCs of different origin might possess a different lineage bias.