UniProt-DAAC: domain architecture alignment and classification, a new method for automatic functional annotation in UniProtKB

Mutual annotation-based prediction of protein domain functions with Domain2GO

Identifying unknown functional properties of proteins is essential for understanding their roles in both health and disease states. The domain composition of a protein can reveal critical information in this context, as domains are structural and functional units that dictate how the protein should act at the molecular level. The expensive and time-consuming nature of wet-lab experimental approaches prompted researchers to develop computational strategies for predicting the functions of proteins. In this study, we proposed a new method called Domain2GO that infers associations between protein domains and function-defining gene ontology (GO) terms, thus redefining the problem as domain function prediction. Domain2GO uses documented protein-level GO annotations together with proteins' domain annotations. Co-annotation patterns of domains and GO terms in the same proteins are examined using statistical resampling to obtain reliable associations. As a use-case study, we evaluated the biological relevance of examples selected from the Domain2GO-generated domain-GO term mappings via literature review. Then, we applied Domain2GO to predict unknown protein functions by propagating domain-associated GO terms to proteins annotated with these domains. For function prediction performance evaluation and comparison against other methods, we employed Critical Assessment of Function Annotation 3 (CAFA3) challenge datasets. The results demonstrated the high potential of Domain2GO, particularly for predicting molecular function and biological process terms, along with advantages such as producing interpretable results and having an exceptionally low computational cost. The approach presented here can be extended to other ontologies and biological entities to investigate unknown relationships in complex and large-scale biological data. The source code, datasets, results, and user instructions for Domain2GO are available at https://github.com/HUBioDataLab/Domain2GO. Additionally, we offer a user-friendly online tool at https://huggingface.co/spaces/HUBioDataLab/Domain2GO, which simplifies the prediction of functions of previously unannotated proteins solely using amino acid sequences.

RNAseq analysis of Cellvibrio japonicus during starch utilization differentiates between genes encoding carbohydrate active enzymes controlled by substrate detection or growth rate

IMPORTANCE

Understanding the bacterial metabolism of starch is important as this polysaccharide is a ubiquitous ingredient in foods, supplements, and medicines, all of which influence gut microbiome composition and health. Our RNAseq and growth data set provides a valuable resource to those who want to better understand the regulation of starch utilization in Gram-negative bacteria. These data are also useful as they provide an example of how to approach studying a starch-utilizing bacterium that has many putative amylases by coupling transcriptomic data with growth assays to overcome the potential challenges of functional redundancy. The RNAseq data can also be used as a part of larger meta-analyses to compare how C. japonicus regulates carbohydrate active enzymes, or how this bacterium compares to gut microbiome constituents in terms of starch utilization potential.

ASCARIS: Positional feature annotation and protein structure-based representation of single amino acid variations

Background

Genomic variations may cause deleterious effects on protein functionality and perturb biological processes. Elucidating the effects of variations is critical for developing novel treatment strategies for diseases of genetic origin. Computational approaches have been aiding the work in this field by modeling and analyzing the mutational landscape. However, new approaches are required, especially for accurate representation and data-centric analysis of sequence variations.

Method

In this study, we propose ASCARIS (Annotation and StruCture-bAsed RepresentatIon of Single amino acid variations), a method for the featurization (i.e., quantitative representation) of single amino acid variations (SAVs), which could be used for a variety of purposes, such as predicting their functional effects or building multi-omics-based integrative models. ASCARIS utilizes the direct and spatial correspondence between the location of the SAV on the sequence/structure and 30 different types of positional feature annotations (e.g., active/lipidation/glycosylation sites; calcium/metal/DNA binding, inter/transmembrane regions, etc.), along with structural features and physicochemical properties. The main novelty of this method lies in constructing reusable numerical representations of SAVs via functional annotations.

Results

We statistically analyzed the relationship between these features and the consequences of variations and found that each carries information in this regard. To investigate potential applications of ASCARIS, we trained variant effect prediction models that utilize our SAV representations as input. We carried out an ablation study and a comparison against the state-of-the-art methods and observed that ASCARIS has a competing and complementary performance against widely-used predictors. ASCARIS can be used alone or in combination with other approaches to represent SAVs from a functional perspective. ASCARIS is available as a programmatic tool at https://github.com/HUBioDataLab/ASCARIS and as a web-service at https://huggingface.co/spaces/HUBioDataLab/ASCARIS.

An O2-sensing diguanylate cyclase broadly affects the aerobic transcriptome in the phytopathogen Pectobacterium carotovorum

Pectobacterium carotovorum is an important plant pathogen responsible for the destruction of crops through bacterial soft rot, which is modulated by oxygen (O₂) concentration. A soluble globin coupled sensor protein, Pcc DgcO (also referred to as PccGCS) is one way through which P. carotovorum senses oxygen. DgcO contains a diguanylate cyclase output domain producing c-di-GMP. Synthesis of the bacterial second messenger c-di-GMP is increased upon oxygen binding to the sensory globin domain. This work seeks to understand regulation of function by DgcO at the transcript level. RNA sequencing and differential expression analysis revealed that the deletion of DgcO only affects transcript levels in cells grown under aerobic conditions. Differential expression analysis showed that DgcO deletion alters transcript levels for metal transporters. These results, followed by inductively coupled plasma-mass spectrometry showing decreased concentrations of six biologically relevant metals upon DgcO deletion, provide evidence that a globin coupled sensor can affect cellular metal content. These findings improve the understanding of the transcript level control of O₂-dependent phenotypes in an important phytopathogen and establish a basis for further studies on c-di-GMP-dependent functions in P. carotovorum.

Clinical and bi-genomic DNA findings of patients suspected to have mitochondrial diseases

Background: Mitochondrial diseases are the most common group of inherited metabolic disorders, causing difficulties in definite diagnosis due to clinical and genetic heterogeneity. Clinical components are predominantly associated with pathogenic variants shown in nuclear or mitochondrial genomes that affect vital respiratory chain function. The development of high-throughput sequencing technologies has accelerated the elucidation of the genetic etiology of many genetic diseases that previously remained undiagnosed. Methods: Thirty affected patients from 24 unrelated families with clinical, radiological, biochemical, and histopathological evaluations considered for mitochondrial diseases were investigated. DNA isolated from the peripheral blood samples of probands was sequenced for nuclear exome and mitochondrial DNA (mtDNA) analyses. MtDNA sequencing was also performed from the muscle biopsy material in one patient. For segregation, Sanger sequencing is performed for pathogenic alterations in five other affected family members and healthy parents. Results: Exome sequencing revealed 14 different pathogenic variants in nine genes encoding mitochondrial function peptides (AARS2, EARS2, ECHS1, FBXL4, MICOS13, NDUFAF6, OXCT1, POLG, and TK2) in 12 patients from nine families and four variants in genes encoding important for muscle structure (CAPN3, DYSF, and TCAP) in six patients from four families. Three probands carried pathogenic mtDNA variations in two genes (MT-ATP6 and MT-TL1). Nine variants in five genes are reported for the first time with disease association: (AARS2: c.277C>T/p.(R93*), c.845C>G/p.(S282C); EARS2: c.319C>T/p.(R107C), c.1283delC/p.(P428Lfs*); ECHS1: c.161G>A/p.(R54His); c.202G>A/p.(E68Lys); NDUFAF6: c.479delA/p.(N162Ifs*27); and OXCT1: c.1370C>T/p.(T457I), c.1173-139G>T/p.(?). Conclusion: Bi-genomic DNA sequencing clarified genetic etiology in 67% (16/24) of the families. Diagnostic utility by mtDNA sequencing in 13% (3/24) and exome sequencing in 54% (13/24) of the families prioritized searching for nuclear genome pathologies for the first-tier test. Weakness and muscle wasting observed in 17% (4/24) of the families underlined that limb-girdle muscular dystrophy, similar to mitochondrial myopathy, is an essential point for differential diagnosis. The correct diagnosis is crucial for comprehensive genetic counseling of families. Also, it contributes to making treatment-helpful referrals, such as ensuring early access to medication for patients with mutations in the TK2 gene.

FAS: assessing the similarity between proteins using multi-layered feature architectures

MOTIVATION

Protein sequence comparison is a fundamental element in the bioinformatics toolkit. When sequences are annotated with features such as functional domains, transmembrane domains, low complexity regions or secondary structure elements, the resulting feature architectures allow better informed comparisons. However, many existing schemes for scoring architecture similarities cannot cope with features arising from multiple annotation sources. Those that do fall short in the resolution of overlapping and redundant feature annotations.

RESULTS

Here, we introduce FAS, a scoring method that integrates features from multiple annotation sources in a directed acyclic architecture graph. Redundancies are resolved as part of the architecture comparison by finding the paths through the graphs that maximize the pair-wise architecture similarity. In a large-scale evaluation on more than 10 000 human-yeast ortholog pairs, architecture similarities assessed with FAS are consistently more plausible than those obtained using e-values to resolve overlaps or leaving overlaps unresolved. Three case studies demonstrate the utility of FAS on architecture comparison tasks: benchmarking of orthology assignment software, identification of functionally diverged orthologs, and diagnosing protein architecture changes stemming from faulty gene predictions. With the help of FAS, feature architecture comparisons can now be routinely integrated into these and many other applications.

AVAILABILITY AND IMPLEMENTATION

FAS is available as python package: https://pypi.org/project/greedyFAS/.

ProFAB-open protein functional annotation benchmark

As the number of protein sequences increases in biological databases, computational methods are required to provide accurate functional annotation with high coverage. Although several machine learning methods have been proposed for this purpose, there are still two main issues: (i) construction of reliable positive and negative training and validation datasets, and (ii) fair evaluation of their performances based on predefined experimental settings. To address these issues, we have developed ProFAB: Open Protein Functional Annotation Benchmark, which is a platform providing an infrastructure for a fair comparison of protein function prediction methods. ProFAB provides filtered and preprocessed protein annotation datasets and enables the training and evaluation of function prediction methods via several options. We believe that ProFAB will be useful for both computational and experimental researchers by enabling the utilization of ready-to-use datasets and machine learning algorithms for protein function prediction based on Gene Ontology terms and Enzyme Commission numbers. ProFAB is available at https://github.com/kansil/ProFAB and https://profab.kansil.org.

How to approach machine learning-based prediction of drug/compound-target interactions

The identification of drug/compound-target interactions (DTIs) constitutes the basis of drug discovery, for which computational predictive approaches have been developed. As a relatively new data-driven paradigm, proteochemometric (PCM) modeling utilizes both protein and compound properties as a pair at the input level and processes them via statistical/machine learning. The representation of input samples (i.e., proteins and their ligands) in the form of quantitative feature vectors is crucial for the extraction of interaction-related properties during the artificial learning and subsequent prediction of DTIs. Lately, the representation learning approach, in which input samples are automatically featurized via training and applying a machine/deep learning model, has been utilized in biomedical sciences. In this study, we performed a comprehensive investigation of different computational approaches/techniques for protein featurization (including both conventional approaches and the novel learned embeddings), data preparation and exploration, machine learning-based modeling, and performance evaluation with the aim of achieving better data representations and more successful learning in DTI prediction. For this, we first constructed realistic and challenging benchmark datasets on small, medium, and large scales to be used as reliable gold standards for specific DTI modeling tasks. We developed and applied a network analysis-based splitting strategy to divide datasets into structurally different training and test folds. Using these datasets together with various featurization methods, we trained and tested DTI prediction models and evaluated their performance from different angles. Our main findings can be summarized under 3 items: (i) random splitting of datasets into train and test folds leads to near-complete data memorization and produce highly over-optimistic results, as a result, should be avoided, (ii) learned protein sequence embeddings work well in DTI prediction and offer high potential, despite interaction-related properties (e.g., structures) of proteins are unused during their self-supervised model training, and (iii) during the learning process, PCM models tend to rely heavily on compound features while partially ignoring protein features, primarily due to the inherent bias in DTI data, indicating the requirement for new and unbiased datasets. We hope this study will aid researchers in designing robust and high-performing data-driven DTI prediction systems that have real-world translational value in drug discovery.

Machine learning-based prediction of drug approvals using molecular, physicochemical, clinical trial, and patent-related features

BACKGROUND

Drug development productivity has been declining lately due to elevated costs and reduced discovery rates. Therefore, pharmaceutical companies have been seeking alternative ways to determine and evaluate drug candidates.

RESEARCH DESIGN AND METHODS

In this work, we proposed a new computational approach to directly predict the regulatory approval of drug candidates, and implemented it as a method called 'DrugApp.' To accomplish this task, we employed multiple types of features including molecular and physicochemical properties of drug candidates, together with clinical trial and patent-related features, which are then processed by random forest classifiers to train our disease group-specific approval prediction models.

RESULTS

Our evaluations indicated DrugApp has a high and robust prediction performance. Within a use-case study, we showed our method can predict phase IV trial drugs that are later withdrawn from the market due to severe side effects. Finally, we used DrugApp models to forecast the approval of drug candidates that are currently in phases I/II/III of clinical trials.

CONCLUSIONS

We hope that our study will aid the research community in terms of evaluating and improving the process of drug development. The datasets, source code, results, and pre-trained models of DrugApp are freely available at https://github.com/HUBioDataLab/DrugApp.

An 85-amino-acid polypeptide from Myrmeleon bore larvae (antlions) homologous to heat shock factor binding protein 1 with antiproliferative activity against MG-63 osteosarcoma cells in vitro

Background

Venomous arthropods have substances in their venom with antiproliferative potential for neoplastic cells.

Objectives

To identify a polypeptide from Myrmeleon bore (antlion) with antiproliferative activity against neoplastic cells, and to elucidate the molecular mechanism of the activity.

Methods

We used gel filtration and ion exchange chromatography to purify a polypeptide with antiproliferative activity against MG-63 human osteosarcoma cells from a proteinaceous extract of antlion. The polypeptide was sequenced and the stability of its antiproliferative activity was tested under a range of conditions in vitro. An 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was used to determine the antiproliferative activity of the polypeptide against the MG-63 osteosarcoma cells and MC3T3-E1 mouse calvarial osteoblasts, which were used as a non-neoplastic control. We used western blotting to compare the levels of expression of heat shock transcription factor 1 (HSF1), heat shock protein 90 (HSP90), cyclin-dependent kinase 4 (CDK4), and protein kinase B alpha (ATK1) in MG-63 osteosarcoma cells and their mouse homologs in MC3T3-E1 osteoblasts after their treatment with the antlion antiproliferative polypeptide (ALAPP).

Results

The 85-amino-acid ALAPP has a 56% sequence identity with the human heat shock factor binding protein 1 (HSBP1). The antiproliferative activity of the polypeptide is relatively insensitive to temperature, pH, and metal ions. ALAPP has a strong concentration-dependent antiproliferative activity against MG-63 osteosarcoma cells compared with its effect on MC3T3-E1 osteoblasts. ALAPP significantly upregulates the expression of HSF1 in MC3T3-EL osteoblasts, but not in MG-63 osteosarcoma. ALAPP significantly downregulated the expression of HSP90, CDK4, and AKT1 expression in MG-63 osteosarcoma, but not in the osteoblasts.

Conclusions

ALAPP has significant antiproliferative activity against MG-63 osteosarcoma cells, but not nonneoplastic MC3T3-E1 osteoblasts. We speculate that non-neoplastic cells may evade the antiproliferative effect of ALAPP by upregulating HSF1 to maintain their HSP90, CDK4, and AKT1 expression at a relatively constant level.

Protein domain-based prediction of drug/compound-target interactions and experimental validation on LIM kinases

Predictive approaches such as virtual screening have been used in drug discovery with the objective of reducing developmental time and costs. Current machine learning and network-based approaches have issues related to generalization, usability, or model interpretability, especially due to the complexity of target proteins' structure/function, and bias in system training datasets. Here, we propose a new method "DRUIDom" (DRUg Interacting Domain prediction) to identify bio-interactions between drug candidate compounds and targets by utilizing the domain modularity of proteins, to overcome problems associated with current approaches. DRUIDom is composed of two methodological steps. First, ligands/compounds are statistically mapped to structural domains of their target proteins, with the aim of identifying their interactions. As such, other proteins containing the same mapped domain or domain pair become new candidate targets for the corresponding compounds. Next, a million-scale dataset of small molecule compounds, including those mapped to domains in the previous step, are clustered based on their molecular similarities, and their domain associations are propagated to other compounds within the same clusters. Experimentally verified bioactivity data points, obtained from public databases, are meticulously filtered to construct datasets of active/interacting and inactive/non-interacting drug/compound-target pairs (~2.9M data points), and used as training data for calculating parameters of compound-domain mappings, which led to 27,032 high-confidence associations between 250 domains and 8,165 compounds, and a finalized output of ~5 million new compound-protein interactions. DRUIDom is experimentally validated by syntheses and bioactivity analyses of compounds predicted to target LIM-kinase proteins, which play critical roles in the regulation of cell motility, cell cycle progression, and differentiation through actin filament dynamics. We showed that LIMK-inhibitor-2 and its derivatives significantly block the cancer cell migration through inhibition of LIMK phosphorylation and the downstream protein cofilin. One of the derivative compounds (LIMKi-2d) was identified as a promising candidate due to its action on resistant Mahlavu liver cancer cells. The results demonstrated that DRUIDom can be exploited to identify drug candidate compounds for intended targets and to predict new target proteins based on the defined compound-domain relationships. Datasets, results, and the source code of DRUIDom are fully-available at: https://github.com/cansyl/DRUIDom.

Exploring the Meta-regulon of the CRP/FNR Family of Global Transcriptional Regulators in a Partial-Nitritation Anammox Microbiome

Microorganisms must respond to environmental changes to survive, often by controlling transcription initiation. Intermittent aeration during wastewater treatment presents a cyclically changing environment to which microorganisms must react. We used an intermittently aerated bioreactor performing partial nitritation and anammox (PNA) to investigate how the microbiome responds to recurring change. Meta-transcriptomic analysis revealed a dramatic disconnect between the relative DNA abundance and gene expression within the metagenome-assembled genomes (MAGs) of community members, suggesting the importance of transcriptional regulation in this microbiome. To explore how community members responded to cyclic aeration via transcriptional regulation, we searched for homologs of the catabolite repressor protein/fumarate and nitrate reductase regulatory protein (CRP/FNR) family of transcription factors (TFs) within the MAGs. Using phylogenetic analyses, evaluation of sequence conservation in important amino acid residues, and prediction of genes regulated by TFs in the MAGs, we identified homologs of the oxygen-sensing FNR in Nitrosomonas and Rhodocyclaceae, nitrogen-sensing dissimilative nitrate respiration regulator that responds to nitrogen species (DNR) in Rhodocyclaceae, and nitrogen-sensing nitrite and nitric oxide reductase regulator that responds to nitrogen species (NnrR) in Nitrospira MAGs. Our data also predict that CRP/FNR homologs in Ignavibacteria, Flavobacteriales, and Saprospiraceae MAGs sense carbon availability. In addition, a CRP/FNR homolog in a Brocadia MAG was most closely related to CRP TFs known to sense carbon sources in well-studied organisms. However, we predict that in autotrophic Brocadia, this TF most likely regulates a diverse set of functions, including a response to stress during the cyclic aerobic/anoxic conditions. Overall, this analysis allowed us to define a meta-regulon of the PNA microbiome that explains functions and interactions of the most active community members.

IMPORTANCE Microbiomes are important contributors to many ecosystems, including ones where nutrient cycling is stimulated by aeration control. Optimizing cyclic aeration helps reduce energy needs and maximize microbiome performance during wastewater treatment; however, little is known about how most microbial community members respond to these alternating conditions. We defined the meta-regulon of a PNA microbiome by combining existing knowledge of how the CRP/FNR family of bacterial TFs respond to stimuli, with metatranscriptomic analyses to characterize gene expression changes during aeration cycles. Our results indicated that, for some members of the community, prior knowledge is sufficient for high-confidence assignments of TF function, whereas other community members have CRP/FNR TFs for which inferences of function are limited by lack of prior knowledge. This study provides a framework to begin elucidating meta-regulons in microbiomes, where pure cultures are not available for traditional transcriptional regulation studies. Defining the meta-regulon can help in optimizing microbiome performance.

Assessment of a complete and classified platelet proteome from genome-wide transcripts of human platelets and megakaryocytes covering platelet functions

Novel platelet and megakaryocyte transcriptome analysis allows prediction of the full or theoretical proteome of a representative human platelet. Here, we integrated the established platelet proteomes from six cohorts of healthy subjects, encompassing 5.2 k proteins, with two novel genome-wide transcriptomes (57.8 k mRNAs). For 14.8 k protein-coding transcripts, we assigned the proteins to 21 UniProt-based classes, based on their preferential intracellular localization and presumed function. This classified transcriptome-proteome profile of platelets revealed: (i) Absence of 37.2 k genome-wide transcripts. (ii) High quantitative similarity of platelet and megakaryocyte transcriptomes (R = 0.75) for 14.8 k protein-coding genes, but not for 3.8 k RNA genes or 1.9 k pseudogenes (R = 0.43–0.54), suggesting redistribution of mRNAs upon platelet shedding from megakaryocytes. (iii) Copy numbers of 3.5 k proteins that were restricted in size by the corresponding transcript levels (iv) Near complete coverage of identified proteins in the relevant transcriptome (log2fpkm > 0.20) except for plasma-derived secretory proteins, pointing to adhesion and uptake of such proteins. (v) Underrepresentation in the identified proteome of nuclear-related, membrane and signaling proteins, as well proteins with low-level transcripts. We then constructed a prediction model, based on protein function, transcript level and (peri)nuclear localization, and calculated the achievable proteome at ~ 10 k proteins. Model validation identified 1.0 k additional proteins in the predicted classes. Network and database analysis revealed the presence of 2.4 k proteins with a possible role in thrombosis and hemostasis, and 138 proteins linked to platelet-related disorders. This genome-wide platelet transcriptome and (non)identified proteome database thus provides a scaffold for discovering the roles of unknown platelet proteins in health and disease.

The molecular basis for the pH-dependent calcium affinity of the pattern recognition receptor langerin

The C-type lectin receptor langerin plays a vital role in the mammalian defense against invading pathogens. Langerin requires a Ca²⁺ cofactor, the binding affinity of which is regulated by pH. Thus, Ca²⁺ is bound when langerin is on the membrane but released when langerin and its pathogen substrate traffic to the acidic endosome, allowing the substrate to be degraded. The change in pH is sensed by protonation of the allosteric pH sensor histidine H294. However, the mechanism by which Ca²⁺ is released from the buried binding site is not clear. We studied the structural consequences of protonating H294 by molecular dynamics simulations (total simulation time: about 120 μs) and Markov models. We discovered a relay mechanism in which a proton is moved into the vicinity of the Ca²⁺-binding site without transferring the initial proton from H294. Protonation of H294 unlocks a conformation in which a protonated lysine side chain forms a hydrogen bond with a Ca²⁺-coordinating aspartic acid. This destabilizes Ca²⁺ in the binding pocket, which we probed by steered molecular dynamics. After Ca²⁺ release, the proton is likely transferred to the aspartic acid and stabilized by a dyad with a nearby glutamic acid, triggering a conformational transition and thus preventing Ca²⁺ rebinding. These results show how pH regulation of a buried orthosteric binding site from a solvent-exposed allosteric pH sensor can be realized by information transfer through a specific chain of conformational arrangements.

Self-assembled fibrinogen–fibronectin hybrid protein nanofibers with medium-sensitive stability

Hybrid protein nanofibers (hPNFs) have been identified as promising nano building blocks for numerous applications in nanomedicine and tissue engineering. We have recently reported a nature-inspired, self-assembly route to create hPNFs from human plasma proteins, i.e., albumin and hemoglobin. However, it is still unclear whether the same route can be applied to other plasma proteins and whether it is possible to control the composition of the resulting fibers. In this context, to further understand the hPNFs self-assembly mechanism and to optimize their properties, we report herein on ethanol-induced self-assembly of two different plasma proteins, i.e., fibrinogen (FG) and fibronectin (FN). We show that by varying initial protein ratios, the composition and thus the properties of the resulting hPNFs can be fine-tuned. Specifically, atomic force microscopy, hydrodynamic diameter, and zeta potential data together revealed a strong correlation of the hPNFs dimensions and surface charge to their initial protein mixing ratio. The composition-independent prompt dissolution of hPNFs in ultrapure water, in contrast to their stability in PBS, indicates that the molecular arrangement of FN and FG in hPNFs is mainly based on electrostatic interactions. Supported by experimental data we introduce a feasible mechanism that explains the interactions between FN and FG and their self-assembly to hPNFs. These findings contribute to the understanding of dual protein interactions, which can be beneficial in designing innovative biomaterials with multifaceted biological and physical characteristics.

Hybrid protein nanofibers (hPNFs) have been identified as promising nano building blocks for numerous applications in nanomedicine and tissue engineering.

Periplasm-enriched fractions from Xanthomonas citri subsp. citri type A and X. fuscans subsp. aurantifolii type B present distinct proteomic profiles under in vitro pathogenicity induction

The causative agent of Asiatic citrus canker, the Gram-negative bacterium Xanthomonas citri subsp. citri (XAC), produces more severe symptoms and attacks a larger number of citric hosts than Xanthomonas fuscans subsp. aurantifolii XauB and XauC, the causative agents of cancrosis, a milder form of the disease. Here we report a comparative proteomic analysis of periplasmic-enriched fractions of XAC and XauB in XAM-M, a pathogenicity- inducing culture medium, for identification of differential proteins. Proteins were resolved by two-dimensional electrophoresis combined with liquid chromatography-mass spectrometry. Among the 12 proteins identified from the 4 unique spots from XAC in XAM-M (p<0.05) were phosphoglucomutase (PGM), enolase, xylose isomerase (XI), transglycosylase, NAD(P)H-dependent glycerol 3-phosphate dehydrogenase, succinyl-CoA synthetase β subunit, 6-phosphogluconate dehydrogenase, and conserved hypothetical proteins XAC0901 and XAC0223; most of them were not detected as differential for XAC when both bacteria were grown in NB medium, a pathogenicity non-inducing medium. XauB showed a very different profile from XAC in XAM-M, presenting 29 unique spots containing proteins related to a great diversity of metabolic pathways. Preponderant expression of PGM and XI in XAC was validated by Western Blot analysis in the periplasmic-enriched fractions of both bacteria. This work shows remarkable differences between the periplasmic-enriched proteomes of XAC and XauB, bacteria that cause symptoms with distinct degrees of severity during citrus infection. The results suggest that some proteins identified in XAC can have an important role in XAC pathogenicity.

Inhibiting Sialidase-Induced TGF-β1 Activation Attenuates Pulmonary Fibrosis in Mice

The active form of transforming growth factor-β1 (TGF-β1) plays a key role in potentiating fibrosis. TGF-β1 is sequestered in an inactive state by a latency-associated glycopeptide (LAP). Sialidases (also called neuraminidases (NEU)) cleave terminal sialic acids from glycoconjugates. The sialidase NEU3 is upregulated in fibrosis, and mice lacking Neu3 show attenuated bleomycin-induced increases in active TGF-β1 in the lungs and attenuated pulmonary fibrosis. Here we observe that recombinant human NEU3 upregulates active human TGF-β1 by releasing active TGF-β1 from its latent inactive form by desialylating LAP. Based on the proposed mechanism of action of NEU3, we hypothesized that compounds with a ring structure resembling picolinic acid might be transition state analogs and thus possible NEU3 inhibitors. Some compounds in this class showed nanomolar IC₅₀ for recombinant human NEU3 releasing active human TGF-β1 from the latent inactive form. The compounds given as daily 0.1-1-mg/kg injections starting at day 10 strongly attenuated lung inflammation, lung TGF-β1 upregulation, and pulmonary fibrosis at day 21 in a mouse bleomycin model of pulmonary fibrosis. These results suggest that NEU3 participates in fibrosis by desialylating LAP and releasing TGF-β1 and that the new class of NEU3 inhibitors are potential therapeutics for fibrosis. SIGNIFICANCE STATEMENT: The extracellular sialidase NEU3 appears to be a key driver of pulmonary fibrosis. The significance of this report is that 1) we show the mechanism (NEU3 desialylates the latency-associated glycopeptide protein that keeps the profibrotic cytokine transforming growth factor-β1 (TGF-β1) in an inactive state, causing active TGF-β1 release), 2) we then use the predicted NEU3 mechanism to identify nM IC₅₀ NEU3 inhibitors, and 3) these new NEU3 inhibitors are potent therapeutics in a mouse model of pulmonary fibrosis.

UniProt: a worldwide hub of protein knowledge

The UniProt Knowledgebase is a collection of sequences and annotations for over 120 million proteins across all branches of life. Detailed annotations extracted from the literature by expert curators have been collected for over half a million of these proteins. These annotations are supplemented by annotations provided by rule based automated systems, and those imported from other resources. In this article we describe significant updates that we have made over the last 2 years to the resource. We have greatly expanded the number of Reference Proteomes that we provide and in particular we have focussed on improving the number of viral Reference Proteomes. The UniProt website has been augmented with new data visualizations for the subcellular localization of proteins as well as their structure and interactions. UniProt resources are available under a CC-BY (4.0) license via the web at https://www.uniprot.org/.

DEEPScreen: high performance drug-target interaction prediction with convolutional neural networks using 2-D structural compound representations

The identification of physical interactions between drug candidate compounds and target biomolecules is an important process in drug discovery. Since conventional screening procedures are expensive and time consuming, computational approaches are employed to provide aid by automatically predicting novel drug-target interactions (DTIs). In this study, we propose a large-scale DTI prediction system, DEEPScreen, for early stage drug discovery, using deep convolutional neural networks. One of the main advantages of DEEPScreen is employing readily available 2-D structural representations of compounds at the input level instead of conventional descriptors that display limited performance. DEEPScreen learns complex features inherently from the 2-D representations, thus producing highly accurate predictions. The DEEPScreen system was trained for 704 target proteins (using curated bioactivity data) and finalized with rigorous hyper-parameter optimization tests. We compared the performance of DEEPScreen against the state-of-the-art on multiple benchmark datasets to indicate the effectiveness of the proposed approach and verified selected novel predictions through molecular docking analysis and literature-based validation. Finally, JAK proteins that were predicted by DEEPScreen as new targets of a well-known drug cladribine were experimentally demonstrated in vitro on cancer cells through STAT3 phosphorylation, which is the downstream effector protein. The DEEPScreen system can be exploited in the fields of drug discovery and repurposing for in silico screening of the chemogenomic space, to provide novel DTIs which can be experimentally pursued. The source code, trained "ready-to-use" prediction models, all datasets and the results of this study are available at ; https://github.com/cansyl/DEEPscreen.

DEEPred: Automated Protein Function Prediction with Multi-task Feed-forward Deep Neural Networks

Automated protein function prediction is critical for the annotation of uncharacterized protein sequences, where accurate prediction methods are still required. Recently, deep learning based methods have outperformed conventional algorithms in computer vision and natural language processing due to the prevention of overfitting and efficient training. Here, we propose DEEPred, a hierarchical stack of multi-task feed-forward deep neural networks, as a solution to Gene Ontology (GO) based protein function prediction. DEEPred was optimized through rigorous hyper-parameter tests, and benchmarked using three types of protein descriptors, training datasets with varying sizes and GO terms form different levels. Furthermore, in order to explore how training with larger but potentially noisy data would change the performance, electronically made GO annotations were also included in the training process. The overall predictive performance of DEEPred was assessed using CAFA2 and CAFA3 challenge datasets, in comparison with the state-of-the-art protein function prediction methods. Finally, we evaluated selected novel annotations produced by DEEPred with a literature-based case study considering the ‘biofilm formation process’ in Pseudomonas aeruginosa. This study reports that deep learning algorithms have significant potential in protein function prediction; particularly when the source data is large. The neural network architecture of DEEPred can also be applied to the prediction of the other types of ontological associations. The source code and all datasets used in this study are available at: https://github.com/cansyl/DEEPred.

Evolution of Protein Domain Architectures

This chapter reviews current research on how protein domain architectures evolve. We begin by summarizing work on the phylogenetic distribution of proteins, as this will directly impact which domain architectures can be formed in different species. Studies relating domain family size to occurrence have shown that they generally follow power law distributions, both within genomes and larger evolutionary groups. These findings were subsequently extended to multi-domain architectures. Genome evolution models that have been suggested to explain the shape of these distributions are reviewed, as well as evidence for selective pressure to expand certain domain families more than others. Each domain has an intrinsic combinatorial propensity, and the effects of this have been studied using measures of domain versatility or promiscuity. Next, we study the principles of protein domain architecture evolution and how these have been inferred from distributions of extant domain arrangements. Following this, we review inferences of ancestral domain architecture and the conclusions concerning domain architecture evolution mechanisms that can be drawn from these. Finally, we examine whether all known cases of a given domain architecture can be assumed to have a single common origin (monophyly) or have evolved convergently (polyphyly). We end by a discussion of some available tools for computational analysis or exploitation of protein domain architectures and their evolution.

ECPred: a tool for the prediction of the enzymatic functions of protein sequences based on the EC nomenclature

BACKGROUND

The automated prediction of the enzymatic functions of uncharacterized proteins is a crucial topic in bioinformatics. Although several methods and tools have been proposed to classify enzymes, most of these studies are limited to specific functional classes and levels of the Enzyme Commission (EC) number hierarchy. Besides, most of the previous methods incorporated only a single input feature type, which limits the applicability to the wide functional space. Here, we proposed a novel enzymatic function prediction tool, ECPred, based on ensemble of machine learning classifiers.

RESULTS

In ECPred, each EC number constituted an individual class and therefore, had an independent learning model. Enzyme vs. non-enzyme classification is incorporated into ECPred along with a hierarchical prediction approach exploiting the tree structure of the EC nomenclature. ECPred provides predictions for 858 EC numbers in total including 6 main classes, 55 subclass classes, 163 sub-subclass classes and 634 substrate classes. The proposed method is tested and compared with the state-of-the-art enzyme function prediction tools by using independent temporal hold-out and no-Pfam datasets constructed during this study.

CONCLUSIONS

ECPred is presented both as a stand-alone and a web based tool to provide probabilistic enzymatic function predictions (at all five levels of EC) for uncharacterized protein sequences. Also, the datasets of this study will be a valuable resource for future benchmarking studies. ECPred is available for download, together with all of the datasets used in this study, at: https://github.com/cansyl/ECPred . ECPred webserver can be accessed through http://cansyl.metu.edu.tr/ECPred.html .

Genome-wide survey of remote homologues for protein domain superfamilies of known structure reveals unequal distribution across structural classes

Domains are the basic building blocks of proteins which can combine to give rise to different domain architectures. Annotation of domains in a sequence is the first step towards understanding the biological function. Since there are a limited number of folds and evolutionarily related proteins have a similar structure, function can be inferred through remote homology. Computational sequence searches were performed for remote homologues on genomes of around ∼160 000 different organisms, starting from nearly 11 000 superfamily queries of known structure. Case studies revealed that most of the associated domains are involved in the same biological process. Using all the proteins predicted to have at least one structural domain, a coverage of 61% of Pfam families was achieved which is higher than the existing methods (43.36% by SIFTS). Taxonomic analysis of the proteins revealed 493 superfamilies in all the major kingdoms of life and a few lateral gene transfers between viruses and cellular organisms. The distribution of remote homologues across different classes, folds and superfamilies was studied and reveals that sequences are unequally distributed across structural classes. Finally, domain architectures were computed for the homologues and these data were compiled for each superfamily and organism.

HPO2GO: prediction of human phenotype ontology term associations for proteins using cross ontology annotation co-occurrences

Analysing the relationships between biomolecules and the genetic diseases is a highly active area of research, where the aim is to identify the genes and their products that cause a particular disease due to functional changes originated from mutations. Biological ontologies are frequently employed in these studies, which provides researchers with extensive opportunities for knowledge discovery through computational data analysis. In this study, a novel approach is proposed for the identification of relationships between biomedical entities by automatically mapping phenotypic abnormality defining HPO terms with biomolecular function defining GO terms, where each association indicates the occurrence of the abnormality due to the loss of the biomolecular function expressed by the corresponding GO term. The proposed HPO2GO mappings were extracted by calculating the frequency of the co-annotations of the terms on the same genes/proteins, using already existing curated HPO and GO annotation sets. This was followed by the filtering of the unreliable mappings that could be observed due to chance, by statistical resampling of the co-occurrence similarity distributions. Furthermore, the biological relevance of the finalized mappings were discussed over selected cases, using the literature. The resulting HPO2GO mappings can be employed in different settings to predict and to analyse novel gene/protein—ontology term—disease relations. As an application of the proposed approach, HPO term—protein associations (i.e., HPO2protein) were predicted. In order to test the predictive performance of the method on a quantitative basis, and to compare it with the state-of-the-art, CAFA2 challenge HPO prediction target protein set was employed. The results of the benchmark indicated the potential of the proposed approach, as HPO2GO performance was among the best (Fmax = 0.35). The automated cross ontology mapping approach developed in this work may be extended to other ontologies as well, to identify unexplored relation patterns at the systemic level. The datasets, results and the source code of HPO2GO are available for download at: https://github.com/cansyl/HPO2GO.

Large-scale automated function prediction of protein sequences and an experimental case study validation on PTEN transcript variants

Recent advances in computing power and machine learning empower functional annotation of protein sequences and their transcript variations. Here, we present an automated prediction system UniGOPred, for GO annotations and a database of GO term predictions for proteomes of several organisms in UniProt Knowledgebase (UniProtKB). UniGOPred provides function predictions for 514 molecular function (MF), 2909 biological process (BP), and 438 cellular component (CC) GO terms for each protein sequence. UniGOPred covers nearly the whole functionality spectrum in Gene Ontology system and it can predict both generic and specific GO terms. UniGOPred was run on CAFA2 challenge target protein sequences and it is categorized within the top 10 best performing methods for the molecular function category. In addition, the performance of UniGOPred is higher compared to the baseline BLAST classifier in all categories of GO. UniGOPred predictions are compared with UniProtKB/TrEMBL database annotations as well. Furthermore, the proposed tool's ability to predict negatively associated GO terms that defines the functions that a protein does not possess, is discussed. UniGOPred annotations were also validated by case studies on PTEN protein variants experimentally and on CHD8 protein variants with literature. UniGOPred protein functional annotation system is available as an open access tool at http://cansyl.metu.edu.tr/UniGOPred.html.

Protein domain architectures provide a fast, efficient and scalable alternative to sequence-based methods for comparative functional genomics

A functional comparative genome analysis is essential to understand the mechanisms underlying bacterial evolution and adaptation. Detection of functional orthologs using standard global sequence similarity methods faces several problems; the need for defining arbitrary acceptance thresholds for similarity and alignment length, lateral gene acquisition and the high computational cost for finding bi-directional best matches at a large scale. We investigated the use of protein domain architectures for large scale functional comparative analysis as an alternative method. The performance of both approaches was assessed through functional comparison of 446 bacterial genomes sampled at different taxonomic levels. We show that protein domain architectures provide a fast and efficient alternative to methods based on sequence similarity to identify groups of functionally equivalent proteins within and across taxonomic boundaries, and it is suitable for large scale comparative analysis. Running both methods in parallel pinpoints potential functional adaptations that may add to bacterial fitness.