DeepMito: accurate prediction of protein sub-mitochondrial localization using convolutional neural networks

The sulfur-related metabolic status of Aspergillus fumigatus during infection reveals cytosolic serine hydroxymethyltransferase as a promising antifungal target

Sulfur metabolism is an essential aspect of fungal physiology and pathogenicity. Fungal sulfur metabolism comprises anabolic and catabolic routes that are not well conserved in mammals, therefore is considered a promising source of prospective novel antifungal targets. To gain insight into Aspergillus fumigatus sulfur-related metabolism during infection, we used a NanoString custom nCounter-TagSet and compared the expression of 68 key metabolic genes in different murine models of invasive pulmonary aspergillosis, at 3 time-points, and under a variety of in vitro conditions. We identified a set of 15 genes that were consistently expressed at higher levels in vivo than in vitro, suggesting that they may be particularly relevant for intrapulmonary growth and thus constitute promising drug targets. Indeed, the role of 5 of the 15 genes has previously been empirically validated, supporting the likelihood that the remaining candidates are relevant. In addition, the analysis of gene expression dynamics at early (16 h), mid (24 h), and late (72 h) time-points uncovered potential disease initiation and progression factors. We further characterized one of the identified genes, encoding the cytosolic serine hydroxymethyltransferase ShmB, and demonstrated that it is an essential gene of A. fumigatus, also required for virulence in a murine model of established pulmonary infection. We further showed that the structure of the ligand-binding pocket of the fungal enzyme differs significantly from its human counterpart, suggesting that specific inhibitors can be designed. Therefore, in vivo transcriptomics is a powerful tool for identifying genes crucial for fungal pathogenicity that may encode promising antifungal target candidates.

Advancing Yarrowia lipolytica sub-organelle engineering with endogenous mitochondrial targeting sequence

OBJECTIVE

The aim of the study was identification and validation of an endogenous mitochondrial targeting signal (MTS) sequence of Yarrowia lipolytica, for efficient compartmentalization of a target protein to mitochondria.

RESULTS

MTS from citrate synthase of Y. lipolytica (YlCISY-MTS) was identified, isolated and fused with green fluorescent protein (GFP) to direct it to the mitochondrial matrix. The efficiency of localization of GFP to mitochondrial matrix with YlCISY-MTS was compared with currently used MTS from Saccharomyces cerevisiae's cytochrome oxidase subunit IV. Confocal microscopy confirmed the targeted and greater GFP localization, underlining the potential of endogenous YlCISY-MTS for mitochondrial engineering in Y. lipolytica. The availability of endogenous MTS will evade the need of codon optimization of S. cerevisiae MTS for mitochondrial engineering in Y. lipolytica. This is the first report of an endogenous MTS of Y. lipolytica.

CONCLUSION

An endogenous MTS of Y. lipolytica has been identified to facilitate the targeted delivery of a protein in the mitochondria enabling future advancements through leveraging the unique subcellular environment for metabolic engineering applications.

How did we get there? AI applications to biological networks and sequences

The rapidly advancing field of artificial intelligence (AI) has transformed numerous scientific domains, including biology, where a vast and complex volume of data is available for analysis. This paper provides a comprehensive overview of the current state of AI-driven methodologies in genomics, proteomics, and systems biology. We discuss how machine learning algorithms, particularly deep learning models, have enhanced the accuracy and efficiency of embedding sequences, motif discovery, and the prediction of gene expression and protein structure. Additionally, we explore the integration of AI in the embedding and analysis of biological networks, including protein-protein interaction networks and multi-layered networks. By leveraging large-scale biological data, AI techniques have enabled unprecedented insights into complex biological processes and disease mechanisms. This work underlines the potential of applying AI to complex biological data, highlighting current applications and suggesting directions for future research to further explore AI in this rapidly evolving field.

Venomous Cargo: Diverse Toxin-Related Proteins Are Associated with Extracellular Vesicles in Parasitoid Wasp Venom

Unusual membrane-bound particles are present in the venom of the parasitoid wasps that parasitize Drosophila melanogaster. These venom particles harbor about 400 proteins and suppress the encapsulation of a wasp egg. Whereas the proteins in the particles of Leptopilina boulardi venom modify host hemocyte properties, those in L. heterotoma kill host hemocytes. The mechanisms underlying this differential effect are not well understood. The proteome of the L. heterotoma venom particles has been described before, but that of L. boulardi has not been similarly examined. Using sequence-based programs, we report the presence of conserved proteins in both proteomes with strong enrichment in the endomembrane and exosomal cell components. Extracellular vesicle markers are present in both proteomes, as are numerous toxins. Both proteomes also contain proteins lacking any annotation. Among these, we identified the proteins with structural similarity to the ADP-ribosyltransferase enzymes involved in bacterial virulence. We propose that invertebrate fluids like parasitoid venom contain functional extracellular vesicles that deliver toxins and virulence factors from a parasite to a host. Furthermore, the presence of such vesicles may not be uncommon in the venom of other animals. An experimental verification of the predicted toxin functions will clarify the cellular mechanisms underlying successful parasitism.

The crystal structure of yeast mitochondrial type pyrophosphatase provides a model to study pathological mutations in its human ortholog

Mutations in human ppa2 gene encoding mitochondrial inorganic pyrophosphatase (PPA2) result in the mitochondria malfunction in heart and brain and lead to early death. In comparison with its cytosolic counterpart, PPA2 of any species is a poorly characterized enzyme with a previously unknown 3D structure. We report here the crystal structure of PPA2 from yeast Ogataea parapolymorpha (OpPPA2), as well as its biochemical characterization. OpPPA2 is a dimer, demonstrating the fold typical for other eukaryotic Family I pyrophosphatases, including the human cytosolic enzyme. Cofactor Mg2+ ions found in OpPPA2 structure have similar coordination to most known Family I pyrophosphatases. Most of the residues associated with the pathological mutations in human PPA2 are conserved in OpPPA2, and their structural context suggests possible explanations for the effects of the mutations on the human enzyme. In this work, the mutant variant of OpPPA2, Met52Val, corresponding to the natural pathogenic variant Met94Val of human PPA2, is characterized. The obtained structural and biochemical data provide a step to understanding the structural basis of PPA2-associated pathologies.

Protein subcellular localization prediction tools

Protein subcellular localization prediction is of great significance in bioinformatics and biological research. Most of the proteins do not have experimentally determined localization information, computational prediction methods and tools have been acting as an active research area for more than two decades now. Knowledge of the subcellular location of a protein provides valuable information about its functionalities, the functioning of the cell, and other possible interactions with proteins. Fast, reliable, and accurate predictors provides platforms to harness the abundance of sequence data to predict subcellular locations accordingly. During the last decade, there has been a considerable amount of research effort aimed at developing subcellular localization predictors. This paper reviews recent subcellular localization prediction tools in the Eukaryotic, Prokaryotic, and Virus-based categories followed by a detailed analysis. Each predictor is discussed based on its main features, strengths, weaknesses, algorithms used, prediction techniques, and analysis. This review is supported by prediction tools taxonomies that highlight their rele- vant area and examples for uncomplicated categorization and ease of understandability. These taxonomies help users find suitable tools according to their needs. Furthermore, recent research gaps and challenges are discussed to cover areas that need the utmost attention. This survey provides an in-depth analysis of the most recent prediction tools to facilitate readers and can be considered a quick guide for researchers to identify and explore the recent literature advancements.

Histone-like transcription factor Hfl1p in Candida albicans harmonizes nuclear and mitochondrial genomic network in regulation of energy metabolism and filamentation development

Candida albicans is an opportunistic fungal pathogen known for surviving in various nutrient-limited conditions within the host and causing infections. Our prior research revealed that Hfl1p, an archaeal histone-like or Hap5-like protein, is linked to mitochondrial ATP generation and yeast-hyphae morphogenesis. However, the specific roles of Hfl1p in these virulence behaviours, through its function in the CBF/NF-Y complex or as a DNA polymerase II subunit, remain unclear. This study explores Hfl1p's diverse functions in energy metabolism and morphogenesis. By combining proteomic analysis and phenotypic evaluations of the hfl1Δ/hfl1Δ mutant with ChIP data, we found that Hfl1p significantly impacts mitochondrial DNA-encoded CI subunits, the tricarboxylic acid (TCA) cycle, and morphogenetic pathways. This influence occurs either independently or alongside other transcription factors recognizing a conserved DNA motif (TAXXTAATTA). These findings emphasize Hfl1p's critical role in linking carbon metabolism and mitochondrial respiration to the yeast-to-filamentous form transition, enhancing our understanding of C. albicans' metabolic adaptability during morphological transition, an important pathogenic trait of this fungus. This could help identify therapeutic targets by disrupting the relationship between energy metabolism and cell morphology in C. albicans.

NOTCH1 mitochondria localization during heart development promotes mitochondrial metabolism and the endothelial-to-mesenchymal transition in mice

Notch signaling activation drives an endothelial-to-mesenchymal transition (EndMT) critical for heart development, although evidence suggests that the reprogramming of endothelial cell metabolism can regulate endothelial function independent of canonical cell signaling. Herein, we investigated the crosstalk between Notch signaling and metabolic reprogramming in the EndMT process. Biochemically, we find that the NOTCH1 intracellular domain (NICD1) localizes to endothelial cell mitochondria, where it interacts with and activates the complex to enhance mitochondrial metabolism. Targeting NICD1 to mitochondria induces more EndMT compared with wild-type NICD1, and small molecule activation of PDH during pregnancy improves the phenotype in a mouse model of congenital heart defect. A NOTCH1 mutation observed in non-syndromic tetralogy of Fallot patients decreases NICD1 mitochondrial localization and subsequent PDH activity in heart tissues. Altogether, our findings demonstrate NICD1 enrichment in mitochondria of the developing mouse heart, which induces EndMT by activating PDH and subsequently improving mitochondrial metabolism.

Deciphering Membrane Proteins Through Deep Learning Models by Revealing Their Locale Within the Cell

Membrane proteins constitute essential biomolecules attached to or integrated into cellular and organelle membranes, playing diverse roles in cellular processes. Their precise localization is crucial for understanding their functions. Existing protein subcellular localization predictors are predominantly trained on globular proteins; their performance diminishes for membrane proteins, explicitly via deep learning models. To address this challenge, the proposed study segregates membrane proteins into three distinct locations, including the plasma membrane, internal membrane, and membrane of the organelle, using deep learning algorithms including recurrent neural networks (RNN) and Long Short-Term Memory (LSTM). A redundancy-curtailed dataset of 3000 proteins from the MemLoci approach is selected for the investigation, along with incorporating pseudo amino acid composition (PseAAC). PseAAC is an exemplary technique for extracting protein information hidden in the amino acid sequences. After extensive testing, the results show that the accuracy for LSTM and RNN is 83.4% and 80.5%, respectively. The results show that the LSTM model outperforms the RNN and is most commonly employed in proteomics.

Artificial targeting of the NEIL1 DNA glycosylase to the mitochondria

The human NEIL1 DNA glycosylase is one of 11 mammalian glycosylases that initiate base excision repair. While substrate preference, catalytic mechanism, and structural information of NEIL1's ordered residues are available, limited information on its subcellular localization, compounded by relatively low endogenous expression levels, have impeded our understanding of NEIL1. Here, we employed a previously developed computational framework to optimize the mitochondrial localization signal of NEIL1, enabling the visualization of its specific targeting to the mitochondrion via confocal microscopy. While we observed clear mitochondrial localization and increased glycosylase/lyase activity in mitochondrial extracts from low-moderate NEIL1 expression, high NEIL1 mitochondrial expression levels proved harmful, potentially leading to cell death.

PEL-PVP: Application of plant vacuolar protein discriminator based on PEFT ESM-2 and bilayer LSTM in an unbalanced dataset

Plant vacuoles, play a crucial role in maintaining cellular stability, adapting to environmental changes, and responding to external pressures. The accurate identification of vacuolar proteins (PVPs) is crucial for understanding the biosynthetic mechanisms of intracellular vacuoles and the adaptive mechanisms of plants. In order to more accurately identify vacuole proteins, this study developed a new predictive model PEL-PVP based on ESM-2. Through this study, the feasibility and effectiveness of using advanced pre-training models and fine-tuning techniques for bioinformatics tasks were demonstrated, providing new methods and ideas for plant vacuolar protein research. In addition, previous datasets for vacuolar proteins were balanced, but imbalance is more closely related to the actual situation. Therefore, this study constructed an imbalanced dataset UB-PVP from the UniProt database，helping the model better adapt to the complexity and uncertainty in real environments, thereby improving the model's generalization ability and practicality. The experimental results show that compared with existing recognition techniques, achieving significant improvements in multiple indicators, with 6.08 %, 13.51 %, 11.9 %, and 5 % improvements in ACC, SP, MCC, and AUC, respectively. The accuracy reaches 94.59 %, significantly higher than the previous best model GraphIdn. This provides an efficient and precise tool for the study of plant vacuole proteins.

Prediction of mitochondrial targeting signals and their cleavage sites

In this chapter we survey prediction tools and computational methods for the prediction of amino acid sequence elements which target proteins to the mitochondria. We will primarily focus on the prediction of N-terminal mitochondrial targeting signals (MTSs) and their N-terminal cleavage sites by mitochondrial peptidases. We first give practical details useful for using and installing some prediction tools. Then we describe procedures for preparing datasets of MTS containing proteins for statistical analysis or development of new prediction methods. Following that we lightly survey some of the computational techniques used by prediction tools. Finally, after discussing some caveats regarding the reliability of such methods to predict the effects of mutations on MTS function; we close with a discussion of possible future directions of computer prediction methods related to mitochondrial proteins.

SLC25A48 influences plasma levels of choline and localizes to the inner mitochondrial membrane

Choline contributes to the biogenesis of methyl groups, neurotransmitters, and cell membranes. Our genome-wide association study (GWAS) of circulating choline in 2228 college students found that alleles in SLC25A48 (rs6596270) influence choline concentrations in men (p = 9.6 × 10-8), but not women. Previously, the subcellular location and function of SLC25A48 were unknown. Using super-resolution immunofluorescence microscopy, we localized SLC25A48 to the inner mitochondrial membrane. Our results suggest that SLC25A48 transports choline across the inner mitochondrial membrane.

Recent Advancements in Subcellular Proteomics: Growing Impact of Organellar Protein Niches on the Understanding of Cell Biology

The mammalian cell is a complex entity, with membrane-bound and membrane-less organelles playing vital roles in regulating cellular homeostasis. Organellar protein niches drive discrete biological processes and cell functions, thus maintaining cell equilibrium. Cellular processes such as signaling, growth, proliferation, motility, and programmed cell death require dynamic protein movements between cell compartments. Aberrant protein localization is associated with a wide range of diseases. Therefore, analyzing the subcellular proteome of the cell can provide a comprehensive overview of cellular biology. With recent advancements in mass spectrometry, imaging technology, computational tools, and deep machine learning algorithms, studies pertaining to subcellular protein localization and their dynamic distributions are gaining momentum. These studies reveal changing interaction networks because of "moonlighting proteins" and serve as a discovery tool for disease network mechanisms. Consequently, this review aims to provide a comprehensive repository for recent advancements in subcellular proteomics subcontexting methods, challenges, and future perspectives for method developers. In summary, subcellular proteomics is crucial to the understanding of the fundamental cellular mechanisms and the associated diseases.

Proteomic profiling of mesenchymal stem cell-derived extracellular vesicles: Impact of isolation methods on protein cargo

Extracellular vesicles (EVs) are nanosized vesicles with a lipid bilayer that are secreted by cells and play a critical role in cell-to-cell communication. Despite the promising reports regarding their diagnostic and therapeutic potential, the utilization of EVs in the clinical setting is limited due to insufficient information about their cargo and a lack of standardization in isolation and analysis methods. Considering protein cargos in EVs as key contributors to their therapeutic potency, we conducted a tandem mass tag (TMT) quantitative proteomics analysis of three subpopulations of mesenchymal stem cell (MSC)-derived EVs obtained through three different isolation techniques: ultracentrifugation (UC), high-speed centrifugation (HS), and ultracentrifugation on sucrose cushion (SU). Subsequently, we checked EV marker expression, size distribution, and morphological characterization, followed by bioinformatic analysis. The bioinformatic analysis of the proteome results revealed that these subpopulations exhibit distinct molecular and functional characteristics. The choice of isolation method impacts the proteome of isolated EVs by isolating different subpopulations of EVs. Specifically, EVs isolated through the high-speed centrifugation (HS) method exhibited a higher abundance of ribosomal and mitochondrial proteins. Functional apoptosis assays comparing isolated mitochondria with EVs isolated through different methods revealed that HS-EVs, but not other EVs, induced early apoptosis in cancer cells. On the other hand, EVs isolated using the sucrose cushion (SU) and ultracentrifugation (UC) methods demonstrated a higher abundance of proteins primarily involved in the immune response, cell-cell interactions and extracellular matrix interactions. Our analyses unveil notable disparities in proteins and associated biological functions among EV subpopulations, underscoring the importance of meticulously selecting isolation methods and resultant EV subpopulations based on the intended application.

Deep proteomics reveals incorporation of unedited proteins into mitochondrial protein complexes in Arabidopsis

The mitochondrial proteome consists of numerous types of proteins which either are encoded and synthesized in the mitochondria, or encoded in the cell nucleus, synthesized in the cytoplasm and imported into the mitochondria. Their synthesis in the mitochondria, but not in the nucleus, relies on the editing of the primary transcripts of their genes at defined sites. Here, we present an in-depth investigation of the mitochondrial proteome of Arabidopsis (Arabidopsis thaliana) and a public online platform for the exploration of the data. For the analysis of our shotgun proteomic data, an Arabidopsis sequence database was created comprising all available protein sequences from the TAIR10 and Araport11 databases, supplemented with sequences of proteins translated from edited and nonedited transcripts of mitochondria. Amino acid sequences derived from partially edited transcripts were also added to analyze proteins encoded by the mitochondrial genome. Proteins were digested in parallel with six different endoproteases to obtain maximum proteome coverage. The resulting peptide fractions were finally analyzed using liquid chromatography coupled to ion mobility spectrometry and tandem mass spectrometry. We generated a "deep mitochondrial proteome" of 4,692 proteins. 1,339 proteins assigned to mitochondria by the SUBA5 database (https://suba.live) accounted for >80% of the total protein mass of our fractions. The coverage of proteins by identified peptides was particularly high compared to single-protease digests, allowing the exploration of differential splicing and RNA editing events at the protein level. We show that proteins translated from nonedited transcripts can be incorporated into native mitoribosomes and the ATP synthase complex. We present a portal for the use of our data, based on "proteomaps" with directly linked protein data. The portal is available at www.proteomeexplorer.de.

Molecular mechanism of Oxr1p mediated disassembly of yeast V-ATPase

The eukaryotic vacuolar H+-ATPase (V-ATPase) is regulated by reversible disassembly into autoinhibited V1-ATPase and Vo proton channel subcomplexes. We recently reported that the TLDc protein Oxr1p induces V-ATPase disassembly in vitro. Whether and how Oxr1p is involved in enzyme disassembly in vivo, however, is not known. Here, using yeast genetics and fluorescence microscopy, we show that Oxr1p is essential for efficient V-ATPase disassembly in the cell. Supporting biochemical and biophysical in vitro experiments show that whereas Oxr1p-driven holoenzyme disassembly can occur in the absence of nucleotides, the presence of ATP greatly accelerates the process. ATP hydrolysis is needed, however, for subsequent release of Oxr1p so that the free V1 can adopt the autoinhibited conformation. Overall, our study unravels the molecular mechanism of Oxr1p-induced disassembly that occurs in vivo as part of the canonical V-ATPase regulation by reversible disassembly.

Four classic "de novo" genes all have plausible homologs and likely evolved from retro-duplicated or pseudogenic sequences

Despite being previously regarded as extremely unlikely, the idea that entirely novel protein-coding genes can emerge from non-coding sequences has gradually become accepted over the past two decades. Examples of "de novo origination", resulting in lineage-specific "orphan" genes, lacking coding orthologs, are now produced every year. However, many are likely cases of duplicates that are difficult to recognize. Here, I re-examine the claims and show that four very well-known examples of genes alleged to have emerged completely "from scratch"- FLJ33706 in humans, Goddard in fruit flies, BSC4 in baker's yeast and AFGP2 in codfish-may have plausible evolutionary ancestors in pre-existing genes. The first two are likely highly diverged retrogenes coding for regulatory proteins that have been misidentified as orphans. The antifreeze glycoprotein, moreover, may not have evolved from repetitive non-genic sequences but, as in several other related cases, from an apolipoprotein that could have become pseudogenized before later being reactivated. These findings detract from various claims made about de novo gene birth and show there has been a tendency not to invest the necessary effort in searching for homologs outside of a very limited syntenic or phylostratigraphic methodology. A robust approach is used for improving detection that draws upon similarities, not just in terms of statistical sequence analysis, but also relating to biochemistry and function, to obviate notable failures to identify homologs.

Prediction of submitochondrial proteins localization based on Gene Ontology

Mitochondria, which are double-membrane bound organelles commonly found in eukaryotic cells, play a fundamental role as sites for cellular energy production. Within the mitochondria, there exist substructures called submitochondria, and specific proteins associated with submitochondria have been implicated in various human diseases. Therefore, comprehending the precise localization of these submitochondrial proteins is of utmost importance. Such knowledge not only aids in unraveling their role in the pathogenesis of diseases but also facilitates the development of therapeutic drugs and diagnostic methods. In this study, we proposed a novel method based on Gene Ontology (GO) to predict the localization of the submitochondrial proteins, called GO-Submito. More specifically, the GO-Submito fine-tuned pre-training Bidirectional Encoder Representations from Transformers models to encode GO annotations into vectors. Subsequently, the Multi-head Attention Mechanism was employed to fuse these encoded vectors of GO annotations, enabling precise localization prediction. Through comprehensive evaluation, our results demonstrated that GO-Submito outperforms existing methods, offering a reliable and efficient tool for precisely localizing submitochondrial proteins.

Disclosing the locale of transmembrane proteins within cellular alcove by machine learning approach: systematic review and meta analysis

Protein subcellular localization is a promising research question in Proteomics and associated fields, including Biological Sciences, Biomedical Engineering, Computational Biology, Bioinformatics, Proteomics, Artificial Intelligence, and Biophysics. However, computational techniques are preferred to explore this attribute for a massive number of proteins. The byproduct of this conjunction yields diversified location identifiers of proteins. These protein subcellular localization identifiers are unique regarding the database used, organisms, Machine Learning Technique, and accuracy. Despite the availability of these identifiers, the majority of the work has been done on the subcellular localization of proteins and, less work has been done specifically on locations of transmembrane proteins. This systematic review accounts for computational techniques implemented on transmembrane protein localization. Moreover, a literature search on PubMed, Science Direct, and IEEE Databases disclosed no systematic review or meta-analysis on the cell's transmembrane protein locale. A Systematic review was formed under the guidelines of PRISMA by using Science Direct, PubMed, and IEEE Databases. Journal publications from 2000 to 2023 were taken into consideration and screened. This review has focused only on computational studies rather than experimental techniques. 1004 studies were reviewed and were categorized as relevant and non-relevant according to inclusion and exclusion criteria. All the screening was done through Endnote after importing citations. This systematic review characterizes the gap in targeting the locale of the transmembrane protein and will aid researchers in exploring its new horizons.Communicated by Ramaswamy H. Sarma.

Identification of plant vacuole proteins by using graph neural network and contact maps

Plant vacuoles are essential organelles in the growth and development of plants, and accurate identification of their proteins is crucial for understanding their biological properties. In this study, we developed a novel model called GraphIdn for the identification of plant vacuole proteins. The model uses SeqVec, a deep representation learning model, to initialize the amino acid sequence. We utilized the AlphaFold2 algorithm to obtain the structural information of corresponding plant vacuole proteins, and then fed the calculated contact maps into a graph convolutional neural network. GraphIdn achieved accuracy values of 88.51% and 89.93% in independent testing and fivefold cross-validation, respectively, outperforming previous state-of-the-art predictors. As far as we know, this is the first model to use predicted protein topology structure graphs to identify plant vacuole proteins. Furthermore, we assessed the effectiveness and generalization capability of our GraphIdn model by applying it to identify and locate peroxisomal proteins, which yielded promising outcomes. The source code and datasets can be accessed at https://github.com/SJNNNN/GraphIdn .

Uncharacterized protein C17orf80 - a novel interactor of human mitochondrial nucleoids

Molecular functions of many human proteins remain unstudied, despite the demonstrated association with diseases or pivotal molecular structures, such as mitochondrial DNA (mtDNA). This small genome is crucial for the proper functioning of mitochondria, the energy-converting organelles. In mammals, mtDNA is arranged into macromolecular complexes called nucleoids that serve as functional stations for its maintenance and expression. Here, we aimed to explore an uncharacterized protein C17orf80, which was previously detected close to the nucleoid components by proximity labelling mass spectrometry. To investigate the subcellular localization and function of C17orf80, we took advantage of immunofluorescence microscopy, interaction proteomics and several biochemical assays. We demonstrate that C17orf80 is a mitochondrial membrane-associated protein that interacts with nucleoids even when mtDNA replication is inhibited. In addition, we show that C17orf80 is not essential for mtDNA maintenance and mitochondrial gene expression in cultured human cells. These results provide a basis for uncovering the molecular function of C17orf80 and the nature of its association with nucleoids, possibly leading to new insights about mtDNA and its expression.

Insulin-Degrading Enzyme Interacts with Mitochondrial Ribosomes and Respiratory Chain Proteins

Insulin-degrading enzyme (IDE) is a highly conserved metalloprotease that is mainly localized in the cytosol. Although IDE can degrade insulin and some other low molecular weight substrates efficiently, its ubiquitous expression suggests additional functions supported by experimental findings, such as a role in stress responses and cellular protein homeostasis. The translation of a long full-length IDE transcript has been reported to result in targeting to mitochondria, but the role of IDE in this compartment is unknown. To obtain initial leads on the function of IDE in mitochondria, we used a proximity biotinylation approach to identify proteins interacting with wild-type and protease-dead IDE targeted to the mitochondrial matrix. We find that IDE interacts with multiple mitochondrial ribosomal proteins as well as with proteins involved in the synthesis and assembly of mitochondrial complex I and IV. The mitochondrial interactomes of wild type and mutant IDE are highly similar and do not reveal any likely proteolytic IDE substrates. We speculate that IDE could adopt similar additional non-proteolytic functions in mitochondria as in the cytosol, acting as a chaperone and contributing to protein homeostasis and stress responses.

MULocDeep web service for protein localization prediction and visualization at subcellular and suborganellar levels

Predicting protein localization and understanding its mechanisms are critical in biology and pathology. In this context, we propose a new web application of MULocDeep with improved performance, result interpretation, and visualization. By transferring the original model into species-specific models, MULocDeep achieved competitive prediction performance at the subcellular level against other state-of-the-art methods. It uniquely provides a comprehensive localization prediction at the suborganellar level. Besides prediction, our web service quantifies the contribution of single amino acids to localization for individual proteins; for a group of proteins, common motifs or potential targeting-related regions can be derived. Furthermore, the visualizations of targeting mechanism analyses can be downloaded for publication-ready figures. The MULocDeep web service is available at https://www.mu-loc.org/.

Identification and Functional Analysis of Two Mitoferrins, CsMIT1 and CsMIT2, Participating in Iron Homeostasis in Cucumber

Mitochondria are one of the major iron sinks in plant cells. Mitochondrial iron accumulation involves the action of ferric reductase oxidases (FRO) and carriers located in the inner mitochondrial membrane. It has been suggested that among these transporters, mitoferrins (mitochondrial iron transporters, MITs) belonging to the mitochondrial carrier family (MCF) function as mitochondrial iron importers. In this study, two cucumber proteins, CsMIT1 and CsMIT2, with high homology to Arabidopsis, rice and yeast MITs were identified and characterized. CsMIT1 and CsMIT2 were expressed in all organs of the two-week-old seedlings. Under Fe-limited conditions as well as Fe excess, the mRNA levels of CsMIT1 and CsMIT2 were altered, suggesting their regulation by iron availability. Analyses using Arabidopsis protoplasts confirmed the mitochondrial localization of cucumber mitoferrins. Expression of CsMIT1 and CsMIT2 restored the growth of the Δmrs3Δmrs4 mutant (defective in mitochondrial Fe transport), but not in mutants sensitive to other heavy metals. Moreover, the altered cytosolic and mitochondrial Fe concentrations, observed in the Δmrs3Δmrs4 strain, were recovered almost to the levels of WT yeast by expressing CsMIT1 or CsMIT2. These results indicate that cucumber proteins are involved in the iron transport from the cytoplasm to the mitochondria.

Automatic recognition of protein subcellular location patterns in single cells from immunofluorescence images based on deep learning

With the improvement of single-cell measurement techniques, there is a growing awareness that individual differences exist among cells, and protein expression distribution can vary across cells in the same tissue or cell line. Pinpointing the protein subcellular locations in single cells is crucial for mapping functional specificity of proteins and studying related diseases. Currently, research about single-cell protein location is still in its infancy, and most studies and databases do not annotate proteins at the cell level. For example, in the human protein atlas database, an immunofluorescence image stained for a particular protein shows multiple cells, but the subcellular location annotation is for the whole image, ignoring intercellular difference. In this study, we used large-scale immunofluorescence images and image-level subcellular locations to develop a deep-learning-based pipeline that could accurately recognize protein localizations in single cells. The pipeline consisted of two deep learning models, i.e. an image-based model and a cell-based model. The former used a multi-instance learning framework to comprehensively model protein distribution in multiple cells in each image, and could give both image-level and cell-level predictions. The latter firstly used clustering and heuristics algorithms to assign pseudo-labels of subcellular locations to the segmented cell images, and then used the pseudo-labels to train a classification model. Finally, the image-based model was fused with the cell-based model at the decision level to obtain the final ensemble model for single-cell prediction. Our experimental results showed that the ensemble model could achieve higher accuracy and robustness on independent test sets than state-of-the-art methods.

Computational Approaches for Peroxisomal Protein Localization

Computational approaches are practical when investigating putative peroxisomal proteins and for sub-peroxisomal protein localization in unknown protein sequences. Nowadays, advancements in computational methods and Machine Learning (ML) can be used to hasten the discovery of novel peroxisomal proteins and can be combined with more established computational methodologies. Here, we explain and list some of the most used tools and methodologies for novel peroxisomal protein detection and localization.

MTSviewer: A database to visualize mitochondrial targeting sequences, cleavage sites, and mutations on protein structures

Mitochondrial dysfunction is implicated in a wide array of human diseases ranging from neurodegenerative disorders to cardiovascular defects. The coordinated localization and import of proteins into mitochondria are essential processes that ensure mitochondrial homeostasis. The localization and import of most mitochondrial proteins are driven by N-terminal mitochondrial targeting sequences (MTS's), which interact with import machinery and are removed by the mitochondrial processing peptidase (MPP). The recent discovery of internal MTS's-those which are distributed throughout a protein and act as import regulators or secondary MPP cleavage sites-has expanded the role of both MTS's and MPP beyond conventional N-terminal regulatory pathways. Still, the global mutational landscape of MTS's remains poorly characterized, both from genetic and structural perspectives. To this end, we have integrated a variety of tools into one harmonized R/Shiny database called MTSviewer (https://neurobioinfo.github.io/MTSvieweR/), which combines MTS predictions, cleavage sites, genetic variants, pathogenicity predictions, and N-terminomics data with structural visualization using AlphaFold models of human and yeast mitochondrial proteomes. Using MTSviewer, we profiled all MTS-containing proteins across human and yeast mitochondrial proteomes and provide multiple case studies to highlight the utility of this database.

Transcriptome Analysis of Protein Kinase MoCK2, which Affects Acetyl-CoA Metabolism and Import of CK2-Interacting Mitochondrial Proteins into Mitochondria in the Rice Blast Fungus Magnaporthe oryzae

The rice pathogen Magnaporthe oryzae causes severe losses to rice production. Previous studies have shown that the protein kinase MoCK2 is essential for pathogenesis, and this ubiquitous eukaryotic protein kinase might affect several processes in the fungus that are needed for infection. To better understand which cellular processes are affected by MoCK2 activity, we performed a detailed transcriptome sequencing analysis of deletions of the MoCK2 b1 and b2 components in relation to the background strain Ku80 and connected this analysis with the abundance of substrates for proteins in a previous pulldown of the essential CKa subunit of CK2 to estimate the effects on proteins directly interacting with CK2. The results showed that MoCK2 seriously affected carbohydrate metabolism, fatty acid metabolism, amino acid metabolism, and the related transporters and reduced acetyl-CoA production. CK2 phosphorylation can affect the folding of proteins and especially the effective formation of protein complexes by intrinsically disordered or mitochondrial import by destabilizing soluble alpha helices. The upregulated genes found in the pulldown of the b1 and b2 mutants indicate that proteins directly interacting with CK2 are compensatorily upregulated depending on their pulldown. A similar correlation was found for mitochondrial proteins. Taken together, the classes of proteins and the changes in regulation in the b1 and b2 mutants suggest that CK2 has a central role in mitochondrial metabolism, secondary metabolism, and reactive oxygen species (ROS) resistance, in addition to its previously suggested role in the formation of new ribosomes, all of which are processes central to efficient nonself responses as innate immunity. IMPORTANCE The protein kinase CK2 is highly expressed and essential for plants, animals, and fungi, affecting fatty acid-related metabolism. In addition, it directly affects the import of essential mitochondrial proteins into mitochondria. These effects mean that CK2 is essential for lipid metabolism and mitochondrial function and, as shown previously, is crucial for making new translation machinery proteins. Taken together, our new results combined with previously reported results indicate that CK2 is an essential protein necessary for the capacities to launch efficient innate immunity responses and withstand the negative effects of such responses necessary for general resistance against invading bacteria and viruses as well as to interact with plants, withstand plant immunity responses, and kill plant cells.

OrganelX web server for sub-peroxisomal and sub-mitochondrial protein localization and peroxisomal target signal detection

We present the OrganelX e-Science Web Server that provides a user-friendly implementation of the In-Pero and In-Mito classifiers for sub-peroxisomal and sub-mitochondrial localization of peroxisomal and mitochondrial proteins and the Is-PTS1 algorithm for detecting and validating potential peroxisomal proteins carrying a PTS1 signal sequence. The OrganelX e-Science Web Server is available at https://organelx.hpc.rug.nl/fasta/.

Two neuronal peptides encoded from a single transcript regulate mitochondrial complex III in Drosophila

Naturally produced peptides (<100 amino acids) are important regulators of physiology, development, and metabolism. Recent studies have predicted that thousands of peptides may be translated from transcripts containing small open-reading frames (smORFs). Here, we describe two peptides in Drosophila encoded by conserved smORFs, Sloth1 and Sloth2. These peptides are translated from the same bicistronic transcript and share sequence similarities, suggesting that they encode paralogs. Yet, Sloth1 and Sloth2 are not functionally redundant, and loss of either peptide causes animal lethality, reduced neuronal function, impaired mitochondrial function, and neurodegeneration. We provide evidence that Sloth1/2 are highly expressed in neurons, imported to mitochondria, and regulate mitochondrial complex III assembly. These results suggest that phenotypic analysis of smORF genes in Drosophila can provide a wealth of information on the biological functions of this poorly characterized class of genes.

A novel deep learning-assisted hybrid network for plasmodium falciparum parasite mitochondrial proteins classification

Plasmodium falciparum is a parasitic protozoan that can cause malaria, which is a deadly disease. Therefore, the accurate identification of malaria parasite mitochondrial proteins is essential for understanding their functions and identifying novel drug targets. For classifying protein sequences, several adaptive statistical techniques have been devised. Despite significant gains, prediction performance is still constrained by the lack of appropriate feature descriptors and learning strategies in current systems. Moreover, good ground truth data is important for Artificial Intelligence (AI)-based models but there is a lack of that data in the literature. Therefore, in this work, we propose a novel hybrid network that combines 1D Convolutional Neural Network (CNN) and Bidirectional Gated Recurrent Unit (BGRU) to classify the malaria parasite mitochondrial proteins. Furthermore, we curate a sequential data that are collected from National Center for Biotechnology Information (NCBI) and UniProtKB/Swiss-Prot proteins databanks to prepare a dataset that can be used by the research community for AI-based algorithms evaluation. We obtain 4204 cases after preprocessing of the collected data and denote this set of proteins as PF4204. Finally, we conduct an ablation study on several conventional and deep models using PF4204 and the benchmark PF2095 datasets. The proposed model 'CNN-BGRU' obtains the accuracy values of 0.9096 and 0.9857 on PF4204 and PF2095 datasets, respectively. In addition, the CNN-BGRU is compared with state-of-the-arts, where the results illustrate that it can extract robust features and identify proteins accurately.

GraphLoc: a graph neural network model for predicting protein subcellular localization from immunohistochemistry images

Motivation

Recognition of protein subcellular distribution patterns and identification of location biomarker proteins in cancer tissues are important for understanding protein functions and related diseases. Immunohistochemical (IHC) images enable visualizing the distribution of proteins at the tissue level, providing an important resource for the protein localization studies. In the past decades, several image-based protein subcellular location prediction methods have been developed, but the prediction accuracies still have much space to improve due to the complexity of protein patterns resulting from multi-label proteins and variation of location patterns across cell types or states.

Results

Here, we propose a multi-label multi-instance model based on deep graph convolutional neural networks, GraphLoc, to recognize protein subcellular location patterns. GraphLoc builds a graph of multiple IHC images for one protein, learns protein-level representations by graph convolutions, and predicts multi-label information by a dynamic threshold method. Our results show that GraphLoc is a promising model for image-based protein subcellular location prediction with model interpretability. Furthermore, we apply GraphLoc to the identification of candidate location biomarkers and potential members for protein networks. A large portion of the predicted results have supporting evidence from the existing literatures and the new candidates also provide guidance for further experimental screening.

Availability

The dataset and code are available at: www.csbio.sjtu.edu.cn/bioinf/GraphLoc.

Supplementary information

Supplementary data are available at Bioinformatics online.

A Novel Artificial Intelligence System in Formulation Dissolution Prediction

Artificial neural network (ANN) techniques are widely used to screen the data and predict the experimental result in pharmaceutical studies. In this study, a novel dissolution result prediction and screen system with a backpropagation network and regression methods was modeled. For this purpose, 21 groups of dissolution data were used to train and verify the ANN model. Based on the design of input data, the related data were still available to train the ANN model when the formulation composition was changed. Two regression methods, the effective data regression method (EDRM) and the reference line regression method (RLRM), make this system predict dissolution results with a high accuracy rate but use less database than the orthogonal experiment. Based on the decision tree, a data screen function is also realized in this system. This ANN model provides a novel drug prediction system with a decrease in time and cost and also easily facilitates the design of new formulation.

Protein Science Meets Artificial Intelligence: A Systematic Review and a Biochemical Meta-Analysis of an Inter-Field

Proteins are some of the most fascinating and challenging molecules in the universe, and they pose a big challenge for artificial intelligence. The implementation of machine learning/AI in protein science gives rise to a world of knowledge adventures in the workhorse of the cell and proteome homeostasis, which are essential for making life possible. This opens up epistemic horizons thanks to a coupling of human tacit-explicit knowledge with machine learning power, the benefits of which are already tangible, such as important advances in protein structure prediction. Moreover, the driving force behind the protein processes of self-organization, adjustment, and fitness requires a space corresponding to gigabytes of life data in its order of magnitude. There are many tasks such as novel protein design, protein folding pathways, and synthetic metabolic routes, as well as protein-aggregation mechanisms, pathogenesis of protein misfolding and disease, and proteostasis networks that are currently unexplored or unrevealed. In this systematic review and biochemical meta-analysis, we aim to contribute to bridging the gap between what we call binomial artificial intelligence (AI) and protein science (PS), a growing research enterprise with exciting and promising biotechnological and biomedical applications. We undertake our task by exploring "the state of the art" in AI and machine learning (ML) applications to protein science in the scientific literature to address some critical research questions in this domain, including What kind of tasks are already explored by ML approaches to protein sciences? What are the most common ML algorithms and databases used? What is the situational diagnostic of the AI-PS inter-field? What do ML processing steps have in common? We also formulate novel questions such as Is it possible to discover what the rules of protein evolution are with the binomial AI-PS? How do protein folding pathways evolve? What are the rules that dictate the folds? What are the minimal nuclear protein structures? How do protein aggregates form and why do they exhibit different toxicities? What are the structural properties of amyloid proteins? How can we design an effective proteostasis network to deal with misfolded proteins? We are a cross-functional group of scientists from several academic disciplines, and we have conducted the systematic review using a variant of the PICO and PRISMA approaches. The search was carried out in four databases (PubMed, Bireme, OVID, and EBSCO Web of Science), resulting in 144 research articles. After three rounds of quality screening, 93 articles were finally selected for further analysis. A summary of our findings is as follows: regarding AI applications, there are mainly four types: 1) genomics, 2) protein structure and function, 3) protein design and evolution, and 4) drug design. In terms of the ML algorithms and databases used, supervised learning was the most common approach (85%). As for the databases used for the ML models, PDB and UniprotKB/Swissprot were the most common ones (21 and 8%, respectively). Moreover, we identified that approximately 63% of the articles organized their results into three steps, which we labeled pre-process, process, and post-process. A few studies combined data from several databases or created their own databases after the pre-process. Our main finding is that, as of today, there are no research road maps serving as guides to address gaps in our knowledge of the AI-PS binomial. All research efforts to collect, integrate multidimensional data features, and then analyze and validate them are, so far, uncoordinated and scattered throughout the scientific literature without a clear epistemic goal or connection between the studies. Therefore, our main contribution to the scientific literature is to offer a road map to help solve problems in drug design, protein structures, design, and function prediction while also presenting the "state of the art" on research in the AI-PS binomial until February 2021. Thus, we pave the way toward future advances in the synthetic redesign of novel proteins and protein networks and artificial metabolic pathways, learning lessons from nature for the welfare of humankind. Many of the novel proteins and metabolic pathways are currently non-existent in nature, nor are they used in the chemical industry or biomedical field.

Unique SMYD5 Structure Revealed by AlphaFold Correlates with Its Functional Divergence

SMYD5 belongs to a special class of protein lysine methyltransferases with an MYND (Myeloid-Nervy-DEAF1) domain inserted into a SET (Suppressor of variegation, Enhancer of Zeste, Trithorax) domain. Despite recent advances in its functional characterization, the lack of the crystal structure has hindered our understanding of the structure-and-function relationships of this most unique member of the SMYD protein family. Here, we demonstrate the reliability of using AlphaFold structures for understanding the structure and function of SMYD5 by comparing the AlphaFold structures to the known crystal structures of SMYD proteins, using an inter-residue distance maps-based metric. We found that the AlphaFold confidence scores are inversely associated with the refined B-factors and can serve as a structural indicator of conformational flexibility. We also found that the N-terminal sequence of SMYD5, predicted to be a mitochondrial targeting signal, contains a novel non-classical nuclear localization signal. This sequence is structurally flexible and does not have a well-defined conformation, which might facilitate its recognition for SMYD5's cytonuclear transport. The structure of SMYD5 is unique in many aspects. The "crab"-like structure with a large negatively charged cleft provides a potential binding site for basic molecules such as protamines. The less positively charged MYND domain is associated with the undetectable DNA-binding ability. The most surprising feature is an incomplete target lysine access channel that lacks the evolutionarily conserved tri-aromatic arrangement, being associated with the low H3/H4 catalytic activity. This study expands our understanding of the SMYD protein family from a classical two-lobed structure to a structure of its own kind, being as a fundamental determinant of its functional divergence.

Recent Advances in the Prediction of Subcellular Localization of Proteins and Related Topics

Prediction of subcellular localization of proteins from their amino acid sequences has a long history in bioinformatics and is still actively developing, incorporating the latest advances in machine learning and proteomics. Notably, deep learning-based methods for natural language processing have made great contributions. Here, we review recent advances in the field as well as its related fields, such as subcellular proteomics and the prediction/recognition of subcellular localization from image data.

Mammalian HEMK1 methylates glutamine residue of the GGQ motif of mitochondrial release factors

Despite limited reports on glutamine methylation, methylated glutamine is found to be highly conserved in a "GGQ" motif in both prokaryotes and eukaryotes. In bacteria, glutamine methylation of peptide chain release factors 1/2 (RF1/2) by the enzyme PrmC is essential for translational termination and transcript recycling. Two PrmC homologs, HEMK1 and HEMK2, are found in mammals. In contrast to those of HEMK2, the biochemical properties and biological significance of HEMK1 remain largely unknown. In this study, we demonstrated that HEMK1 is an active methyltransferase for the glutamine residue of the GGQ motif of all four putative mitochondrial release factors (mtRFs)-MTRF1, MTRF1L, MRPL58, and MTRFR. In HEMK1-deficient HeLa cells, GGQ motif glutamine methylation was absent in all the mtRFs. We examined cell growth and mitochondrial properties, but disruption of the HEMK1 gene had no considerable impact on the overall cell growth, mtDNA copy number, mitochondrial membrane potential, and mitochondrial protein synthesis under regular culture condition with glucose as a carbon source. Furthermore, cell growth potential of HEMK1 KO cells was still maintained in the respiratory condition with galactose medium. Our results suggest that HEMK1 mediates the GGQ methylation of all four mtRFs in human cells; however, this specific modification seems mostly dispensable in cell growth and mitochondrial protein homeostasis at least for HeLa cells under fermentative culture condition.

Learning protein subcellular localization multi-view patterns from heterogeneous data of imaging, sequence and networks

Location proteomics seeks to provide automated high-resolution descriptions of protein location patterns within cells. Many efforts have been undertaken in location proteomics over the past decades, thereby producing plenty of automated predictors for protein subcellular localization. However, most of these predictors are trained solely from high-throughput microscopic images or protein amino acid sequences alone. Unifying heterogeneous protein data sources has yet to be exploited. In this paper, we present a pipeline called sequence, image, network-based protein subcellular locator (SIN-Locator) that constructs a multi-view description of proteins by integrating multiple data types including images of protein expression in cells or tissues, amino acid sequences and protein-protein interaction networks, to classify the patterns of protein subcellular locations. Proteins were encoded by both handcrafted features and deep learning features, and multiple combining methods were implemented. Our experimental results indicated that optimal integrations can considerately enhance the classification accuracy, and the utility of SIN-Locator has been demonstrated through applying to new released proteins in the human protein atlas. Furthermore, we also investigate the contribution of different data sources and influence of partial absence of data. This work is anticipated to provide clues for reconciliation and combination of multi-source data for protein location analysis.

Characterization of Helianthus annuus Lipoic Acid Biosynthesis: The Mitochondrial Octanoyltransferase and Lipoyl Synthase Enzyme System

Lipoic acid (LA, 6,8-dithiooctanoic acid) is a sulfur containing coenzyme essential for the activity of several key enzymes involved in oxidative and single carbon metabolism in most bacteria and eukaryotes. LA is synthetized by the concerted activity of the octanoyltransferase (LIP2, EC 2.3.1.181) and lipoyl synthase (LIP1, EC 2.8.1.8) enzymes. In plants, pyruvate dehydrogenase (PDH), 2-oxoglutarate dehydrogenase or glycine decarboxylase are essential complexes that need to be lipoylated. These lipoylated enzymes and complexes are located in the mitochondria, while PDH is also present in plastids where it provides acetyl-CoA for de novo fatty acid biosynthesis. As such, lipoylation of PDH could regulate fatty acid synthesis in both these organelles. In the present work, the sunflower LIP1 and LIP2 genes (HaLIP1m and HaLIP2m) were isolated sequenced, cloned, and characterized, evaluating their putative mitochondrial location. The expression of these genes was studied in different tissues and protein docking was modeled. The genes were also expressed in Escherichia coli and Arabidopsis thaliana, where their impact on fatty acid and glycerolipid composition was assessed. Lipidomic studies in Arabidopsis revealed lipid remodeling in lines overexpressing these enzymes and the involvement of both sunflower proteins in the phenotypes observed is discussed in the light of the results obtained.

Computational methods for protein localization prediction

The accurate annotation of protein localization is crucial in understanding protein function in tandem with a broad range of applications such as pathological analysis and drug design. Since most proteins do not have experimentally-determined localization information, the computational prediction of protein localization has been an active research area for more than two decades. In particular, recent machine-learning advancements have fueled the development of new methods in protein localization prediction. In this review paper, we first categorize the main features and algorithms used for protein localization prediction. Then, we summarize a list of protein localization prediction tools in terms of their coverage, characteristics, and accessibility to help users find suitable tools based on their needs. Next, we evaluate some of these tools on a benchmark dataset. Finally, we provide an outlook on the future exploration of protein localization methods.

Production and structure prediction of amylases from Chlorella vulgaris

Amylases are enzymes required for starch degradation and are naturally produced by many microorganisms. These enzymes are used in several fields such as food processing, beverage, and medicine as well as in the formulation of enzymatic detergents proving their significance in modern biotechnology. In this study, a three-stage growth mode was applied to enhance starch production and amylase detection from Chlorella vulgaris. Stress conditions applied in the second stage of cultivation led to an accumulation of proteins (75% DW) and starch (21% DW) and a decrease in biomass. Amylase activities were detected and they showed high production levels especially on day 3 (35 U/ml) and day 5 (22.5 U/ml) of the second and third stages, respectively. The bioinformatic tools used to seek amylase protein sequences from TSA database of C. vulgaris revealed 7 putative genes encoding for 4 α-amylases, 2 β-amylases, and 1 isoamylase. An in silico investigation showed that these proteins are different in their lengths as well as in their cellular localizations and oligomeric states though they share common features like CSRs of GH13 family or active site of GH14 family. In brief, this study allowed for the production and in silico characterization of amylases from C. vulgaris.

RNALocate v2.0: an updated resource for RNA subcellular localization with increased coverage and annotation

Resolving the spatial distribution of the transcriptome at a subcellular level can increase our understanding of biology and diseases. To facilitate studies of biological functions and molecular mechanisms in the transcriptome, we updated RNALocate, a resource for RNA subcellular localization analysis that is freely accessible at http://www.rnalocate.org/ or http://www.rna-society.org/rnalocate/. Compared to RNALocate v1.0, the new features in version 2.0 include (i) expansion of the data sources and the coverage of species; (ii) incorporation and integration of RNA-seq datasets containing information about subcellular localization; (iii) addition and reorganization of RNA information (RNA subcellular localization conditions and descriptive figures for method, RNA homology information, RNA interaction and ncRNA disease information) and (iv) three additional prediction tools: DM3Loc, iLoc-lncRNA and iLoc-mRNA. Overall, RNALocate v2.0 provides a comprehensive RNA subcellular localization resource for researchers to deconvolute the highly complex architecture of the cell.

MULocDeep: A deep-learning framework for protein subcellular and suborganellar localization prediction with residue-level interpretation

Prediction of protein localization plays an important role in understanding protein function and mechanisms. In this paper, we propose a general deep learning-based localization prediction framework, MULocDeep, which can predict multiple localizations of a protein at both subcellular and suborganellar levels. We collected a dataset with 44 suborganellar localization annotations in 10 major subcellular compartments—the most comprehensive suborganelle localization dataset to date. We also experimentally generated an independent dataset of mitochondrial proteins in Arabidopsis thaliana cell cultures, Solanum tuberosum tubers, and Vicia faba roots and made this dataset publicly available. Evaluations using the above datasets show that overall, MULocDeep outperforms other major methods at both subcellular and suborganellar levels. Furthermore, MULocDeep assesses each amino acid’s contribution to localization, which provides insights into the mechanism of protein sorting and localization motifs. A web server can be accessed at http://mu-loc.org.

Mitochondrially localized MPZL3 emerges as a signaling hub of mammalian physiology

MPZL3 is a nuclear-encoded, mitochondrially localized, immunoglobulin-like V-type protein that functions as a key regulator of epithelial cell differentiation, lipid metabolism, ROS production, glycemic control, and energy expenditure. Recently, MPZL3 has surfaced as an important modulator of sebaceous gland function and of hair follicle cycling, an organ transformation process that is also governed by peripheral clock gene activity and PPARγ. Given the phenotype similarities and differences between Mpzl3 and Pparγ knockout mice, we propose that MPZL3 serves as a signaling hub that is regulated by core clock gene products and/or PPARγ to translate signals from these nuclear transcription factors to the mitochondria to modulate circadian and metabolic regulation. Conservation between murine and human MPZL3 suggests that human MPZL3 may have similarly complex functions in health and disease. We summarize current knowledge and discuss future directions to elucidate the full spectrum of MPZL3 functions in mammalian physiology.

Multiple Losses of MSH1, Gain of mtMutS, and Other Changes in the MutS Family of DNA Repair Proteins in Animals

MutS is a key component of the mismatch repair (MMR) pathway. Members of the MutS protein family are present in prokaryotes, eukaryotes, and viruses. Six MutS homologs (MSH1-6) have been identified in yeast, of which three function in nuclear MMR, while MSH1 functions in mitochondrial DNA repair. MSH proteins are believed to be well conserved in animals, except for MSH1-which is thought to be lost. Two intriguing exceptions to this general picture have been found, both in the class Anthozoa within the phylum Cnidaria. First, an ortholog of the yeast-MSH1 was reported in one hexacoral species. Second, a MutS homolog (mtMutS) has been found in the mitochondrial genome of all octocorals. To understand the origin and potential functional implications of these exceptions, we investigated the evolution of the MutS family both in Cnidaria and in animals in general. Our study confirmed the acquisition of octocoral mtMutS by horizontal gene transfer from a giant virus. Surprisingly, we identified MSH1 in all hexacorals and several sponges and placozoans. By contrast, MSH1 orthologs were lacking in other cnidarians, ctenophores, and bilaterian animals. Furthermore, while we identified MSH2 and MSH6 in nearly all animals, MSH4, MSH5, and, especially, MSH3 were missing in multiple species. Overall, our analysis revealed a dynamic evolution of the MutS family in animals, with multiple losses of MSH1, MSH3, some losses of MSH4 and MSH5, and a gain of the octocoral mtMutS. We propose that octocoral mtMutS functionally replaced MSH1 that was present in the common ancestor of Anthozoa.

Computer-Aided Prediction of Protein Mitochondrial Localization

Protein sequences, directly translated from genomic data, need functional and structural annotation. Together with molecular function and biological process, subcellular localization is an important feature necessary for understanding the protein role and the compartment where the mature protein is active. In the case of mitochondrial proteins, their precursor sequences translated by the ribosome machinery include specific patterns from which it is possible not only to recognize their final destination within the organelle but also which of the mitochondrial subcompartments the protein is intended for. Four compartments are routinely discriminated, including the inner and the outer membranes, the intermembrane space, and the matrix. Here we discuss to which extent it is feasible to develop computational methods for detecting mitochondrial targeting peptides in the precursor sequence and to discriminate their final destination in the organelle. We benchmark two of our methods on the general task of recognizing human mitochondrial proteins endowed with an experimentally characterized targeting peptide (TPpred3) and predicting which submitochondrial compartment is the final destination (DeepMito). We describe how to adopt our web servers in order to discriminate which human proteins are endowed with mitochondrial targeting peptides, the position of cleavage sites, and which submitochondrial compartment are intended for. By this, we add some other 1788 human proteins to the 450 ones already manually annotated in UniProt with a mitochondrial targeting peptide, providing for each of them also the characterization of the suborganellar localization.

iDeepSubMito: identification of protein submitochondrial localization with deep learning

Mitochondria are membrane-bound organelles containing over 1000 different proteins involved in mitochondrial function, gene expression and metabolic processes. Accurate localization of those proteins in the mitochondrial compartments is critical to their operation. A few computational methods have been developed for predicting submitochondrial localization from the protein sequences. Unfortunately, most of these computational methods focus on employing biological features or evolutionary information to extract sequence features, which greatly limits the performance of subsequent identification. Moreover, the efficiency of most computational models is still under explored, especially the deep learning feature, which is promising but requires improvement. To address these limitations, we propose a novel computational method called iDeepSubMito to predict the location of mitochondrial proteins to the submitochondrial compartments. First, we adopted a coding scheme using the ProteinELMo to model the probability distribution over the protein sequences and then represent the protein sequences as continuous vectors. Then, we proposed and implemented convolutional neural network architecture based on the bidirectional LSTM with self-attention mechanism, to effectively explore the contextual information and protein sequence semantic features. To demonstrate the effectiveness of our proposed iDeepSubMito, we performed cross-validation on two datasets containing 424 proteins and 570 proteins respectively, and consisting of four different mitochondrial compartments (matrix, inner membrane, outer membrane and intermembrane regions). Experimental results revealed that our method outperformed other computational methods. In addition, we tested iDeepSubMito on the M187, M983 and MitoCarta3.0 to further verify the efficiency of our method. Finally, the motif analysis and the interpretability analysis were conducted to reveal novel insights into subcellular biological functions of mitochondrial proteins. iDeepSubMito source code is available on GitHub at https://github.com/houzl3416/iDeepSubMito.

Prediction of Protein Solubility Based on Sequence Feature Fusion and DDcCNN

BACKGROUND

Prediction of protein solubility is an indispensable prerequisite for pharmaceutical research and production. The general and specific objective of this work is to design a new model for predicting protein solubility by using protein sequence feature fusion and deep dual-channel convolutional neural networks (DDcCNN) to improve the performance of existing prediction models.

METHODS

The redundancy of raw protein is reduced by CD-HIT. The four subsequences are built from protein sequence: one global and three locals. The global subsequence is the entire protein sequence, and these local subsequences are obtained by moving a sliding window with some rules. Using G-gap to extract the features of the above four subsequences, a mixed matrix is constructed as the input of one channel which is composed of three-layer convolutional operating. Additional features are extracted by SCRATCH tool as input of another channel, which is consist of a single convolution in order to find hidden relationships and improve the accuracy of predictor. The outputs of two parallel channels are concatenated as the input of the hidden layer. And the prediction of protein solubility is obtained in the output layer. The best protein solubility prediction model is obtained by doing some comparative experiments of different frameworks.

RESULTS

The performance indicators of DDcCNN model (our designed) are as follows: accuracy of 77.82%, Matthew's correlation coefficient of 0.57, sensitivity of 76.13% and specificity of 79.32%. The results of some comparative experiments show that the overall performance of DDcCNN model is better than existing models (GCNN, LCNN and PCNN). The related models and data are publicly deposited at http://www.ddccnn.wang .

CONCLUSION

The satisfactory performance of DDcCNN model reveals that these features and flexible computational methodologies can reinforce the existing prediction models for better prediction of protein solubility could be applied in several applications, such as to preselect initial targets that are soluble or to alter solubility of target proteins, thus can help to reduce the production cost.