Point cloud upsampling algorithms can improve the resolution of point clouds and generate dense and uniform point clouds, and are an important image processing technology. Significant progress has been made in point cloud upsampling research in recent years. This paper provides a comprehensive survey of point cloud upsampling algorithms. We classify existing point cloud upsampling algorithms into optimization-based methods and deep learning-based methods, and analyze the advantages and limitations of different algorithms from a modular perspective. In addition, we cover some other important issues such as public datasets and performance evaluation metrics. Finally, we conclude this survey by highlighting several future research directions and open issues that should be further addressed.

The major facilitator superfamily (MFS) is the largest secondary transporter family and is responsible for transporting a broad range of substrates across the biomembrane. These proteins are involved in a series of conformational changes during substrate transport. To decipher the transport mechanism, it is necessary to obtain structures of these different conformations. At present, great progress has been made in predicting protein structure based on coevolutionary information. In this study, AlphaFold2 was used to predict different conformational structures for 69 MFS transporters of E. coli after the selective mutation of residues at the interface between the N- and C-terminal domains. The predicted structures for these mutants had small RMSD values when compared to structures obtained using X-ray crystallography, which indicates that AlphaFold2 predicts the structure of MSF transporters with high accuracy. In addition, different conformations of other transporter family proteins have been successfully predicted based on mutation methods. This study provides a structural basis to study the transporting mechanism of the MFS transporters and a method to probe dynamic conformation changes of transporter family proteins when performing their function.

Protein crystallization is the bottleneck in macromolecular crystallography, and crystal recognition is a very important step in the experiment. To improve the recognition accuracy by image classification algorithms further, the Mask R-CNN model is introduced for the detection of protein crystals in this paper. Because the protein crystal image is greatly affected by backlight and precipitate, the contrast limit adaptive histogram equalization (CLAHE) is applied with Mask R-CNN. Meanwhile, the Transfer Learning method is used to optimize the parameters in Mask R-CNN. Through the comparison experiments between this combined algorithm and the original algorithm, it shows that the improved algorithm can effectively improve the accuracy of segmentation.

Nucleoside homeostasis, which is mediated by transporters and channels, is essential for all life on Earth. In Escherichia coli, NupG mediates the transport of nucleosides and was deemed to be the prototype of the nucleoside proton symporter (NHS) family as well as the major facilitator superfamily (MFS). To date, the substrate recognition and transport mechanisms of NHS transporters are still elusive. Here, we report two crystal structures of NupG (wild-type and D323A NupG) resolved at 3.0 Å. Both structures reveal an identical inward-open conformation. Together with molecular docking and molecular dynamics simulations and in vitro uridine-binding assays, we found that the uridine binding site, which locates in the central cavity between N and C domains of NupG, is constituted by R136, T140, F143, Q225, N228, Q261, E264, Y318, and F322. Moreover, we found that D323 is very important for substrate binding via in vitro uridine-binding assays using D323 mutations, even though it does not have direct contact with uridine. Our structural and biochemical data therefore provide an important framework for the mechanistic understanding of nucleoside transporters of the NHS family.