Improving Subcellular Protein Location Classification by Incorporating Three-Dimensional Structure Information

The subcellular locations of proteins are closely related to their functions. In the past few decades, the application of machine learning algorithms to predict subcellular protein locations has been an important topic in proteomics. However, most studies in this field used only amino acid sequences as the data source. Only a few works focused on other protein data types. For example, three-dimensional structures, which contain far more functional protein information than sequences, remain to be explored. In this work, we extracted various handcrafted features to describe the protein structures from physical, chemical, and topological aspects, as well as the learned features obtained by deep neural networks. We then used these features to classify the subcellular protein locations. Our experimental results demonstrated that some of these structural features have a certain effect on the protein location classification, and can help improve the performance of sequence-based location predictors. Our method provides a new view for the analysis of protein spatial distribution, and is anticipated to be used in revealing the relationships between protein structures and functions.

Immunogenic potential of neopeptides depends on parent protein subcellular location

Antigen presentation via the major histocompatibility complex (MHC) is essential for anti-tumor immunity, however the rules that determine what tumor-derived peptides will be immunogenic are still incompletely understood. Here we investigate whether protein subcellular location driven constraints on accessibility of peptides to the MHC associate with potential for peptide immunogenicity. Analyzing over 380,000 of peptides from studies of MHC presentation and peptide immunogenicity, we find clear spatial biases in both eluted and immunogenic peptides. We find that including parent protein location improves prediction of peptide immunogenicity in multiple datasets. In human immunotherapy cohorts, location was associated with response to a neoantigen vaccine, and immune checkpoint blockade responders generally had a higher burden of neopeptides from accessible locations. We conclude that protein subcellular location adds important information for optimizing immunotherapies.

Genetically encoded X-ray cellular imaging for nanoscale protein localization

Spatial resolution defines the physical limit of microscopes for probing biomolecular localization and interactions in cells. Whereas synchrotron-based X-ray microscopy (XRM) represents a unique approach for imaging a whole cell with nanoscale resolution due to its intrinsic nanoscale resolution and great penetration ability, existing approaches to label biomolecules rely on the use of exogenous tags that are multi-step and error-prone. Here, we repurpose engineered peroxidases as genetically encoded X-ray-sensitive tags (GXET) for site-specific labeling of protein-of-interest in mammalian cells. We find that 3,3′-diaminobenzidine (DAB) polymers that are in-situ catalytically formed by fusion-expressed peroxidases are visible under XRM. Using this new tag, we imaged the protein location associated with the alteration of a DNA-methylation pathway with an ultra-high resolution of 30 nanometers. Importantly, the excellent energy resolution of XRM enables multicolor imaging using different peroxidase tags. The development of GXET enlightens the way to nanoscopic imaging for biological studies.

The genes regulating maintenance of cellular protein location are differentially expressed in porcine epithelial oviductal cells during longterm in vitro cultivation

The oviduct is a part of female reproductive tract that is essential for successful fertilization and early embryo development. It is lined with epithelium consisting of two types of cells: ciliated and secretory. The primary function of ciliated oviductal epithelial cells (OECs) is to support the transport of gametes and embryos through the ovary, whereas secretory OECs produce components of the oviductal fluid. Undoubtedly, the oviductal epithelium plays a major part in the early aspects of pregnancy development, from providing an optimal environment for gametes and embryos to supporting fertilization. Therefore, our aim was to gain a better insight into the genetic changes underlying function of these cells. We have harvested OECs from crossbred gilts (n=45), at the age of about nine months and which displayed two regular estrous cycles, and established long-term primary culture of porcine OECs. Microarray analysis was utilized to determine differentially expressed genes during day 1, 7, 15 and 30 of cultivation, with our results revealing54 differentially expressed genes belonging to three ontology groups: „maintenance of location”, „maintenance of protein location” and „maintenance of protein location in cell”. Since the biochemistry and morphology of epithelial cells may change during long term cultivation, we conclude that our results are a reflection of these changes and help to shed a light on porcine OECs properties in in vitro environment.Running title: Maintenance of cellular protein location in porcine epithelial oviductal cells

A Brief Survey of Machine Learning Methods in Protein Sub-Golgi Localization

Background:The location of proteins in a cell can provide important clues to their functions in various biological processes. Thus, the application of machine learning method in the prediction of protein subcellular localization has become a hotspot in bioinformatics. As one of key organelles, the Golgi apparatus is in charge of protein storage, package, and distribution.Objective:The identification of protein location in Golgi apparatus will provide in-depth insights into their functions. Thus, the machine learning-based method of predicting protein location in Golgi apparatus has been extensively explored. The development of protein sub-Golgi apparatus localization prediction should be reviewed for providing a whole background for the fields.Method:The benchmark dataset, feature extraction, machine learning method and published results were summarized.Results:We briefly introduced the recent progresses in protein sub-Golgi apparatus localization prediction using machine learning methods and discussed their advantages and disadvantages.Conclusion:We pointed out the perspective of machine learning methods in protein sub-Golgi localization prediction.

Marine-freshwater prokaryotic transitions require extensive changes in the proteome

The comparison of microbial genomes found in either freshwater or marine habitats indicated that in some cases (SynechococcusandCa. Pelagibacter) there were notable differences in the global isoelectric point (pI) of proteins. We have analysed global metagenomic proteomes and have added more prokaryotes to extend the pI comparison. Without exception, in a set that included archaea and multiple bacterial phyla, the proteome pI distribution varied, with more acidic values in marine and neutral/basic in freshwater microbes. Four pairs of highly related prokaryotes of marine and freshwater origin revealed marked differences manifested mostly in the residues located at the protein surface. This study has also shown that the magnitude of the change depended on protein location (secreted > cytoplasmic > transmembrane) and affected proteins encoded at both core and flexible genome. Our results point to a very extensive variation taking place in microbes when they move from marine (salt-rich) to freshwater habitats. These adaptations would require long evolutionary times to produce changes involving many genes in the core genome. They also point to significant differences in the physiology, probably at the level of membrane functioning, bioenergetics, intracellular ion concentration and pH (or all of them).

FerriTag: A Genetically-Encoded Inducible Tag for Correlative Light-Electron Microscopy

A current challenge is to develop tags to precisely visualize proteins in cells by light and electron microscopy. Here, we introduce FerriTag, a genetically-encoded chemically-inducible tag for correlative light-electron microscopy (CLEM). FerriTag is a fluorescent recombinant electron-dense ferritin particle that can be attached to a protein-of-interest using rapamycin-induced heterodimerization. We demonstrate the utility of FerriTag for CLEM by labeling proteins associated with various intracellular structures including mitochondria, plasma membrane, and clathrin-coated pits and vesicles. FerriTagging has a high signal-to-noise ratio and a labeling resolution of 10 ± 5 nm. We demonstrate how FerriTagging allows nanoscale mapping of protein location relative to a subcellular structure, and use it to detail the distribution of huntingtin-interacting protein 1 related (HIP1R) in clathrin-coated pits.

Synthetic physical interactions map kinetochore regulators and regions sensitive to constitutive Cdc14 localization

The location of proteins within eukaryotic cells is often critical for their function and relocation of proteins forms the mainstay of regulatory pathways. To assess the importance of protein location to cellular homeostasis, we have developed a methodology to systematically create binary physical interactions between a query protein and most other members of the proteome. This method allows us to rapidly assess which of the thousands of possible protein interactions modify a phenotype. As proof of principle we studied the kinetochore, a multiprotein assembly that links centromeres to the microtubules of the spindle during cell division. In budding yeast, the kinetochores from the 16 chromosomes cluster together to a single location within the nucleus. The many proteins that make up the kinetochore are regulated through ubiquitylation and phosphorylation. By systematically associating members of the proteome to the kinetochore, we determine which fusions affect its normal function. We identify a number of candidate kinetochore regulators, including the phosphatase Cdc14. We examine where within the kinetochore Cdc14 can act and show that the effect is limited to regions that correlate with known phosphorylation sites, demonstrating the importance of serine phospho-regulation for normal kinetochore homeostasis.