Context-informed subgraph foundation models enable interpretable protein-function prediction.

Proteins are the main, building cell blocks, responsible for the existing cell biological processes. Therefore, precise knowledge of protein function is of great significance. There are a lot of methods which are used for protein comparison and for determining protein function. Some of them use structure alignment, others use sequence alignment, while some use protein descriptors. Here, we use two protein descriptors: Voxel and Ray-based descriptors to encode the structural and biological features of proteins. In biology there is a trend to hierarchically organize the things, like protein functions, cell components and the whole living world. There are a lot of classification systems which generate proteins in tree structure. However, due to the fact that it often happens that one protein has more than one parent, the Directed Acyclic Graph (DAG) hierarchy is used. Gene Ontology (GO) is a system for structural and hierarchical representation of proteins and gene products which support DAG hierarchy. CLUS, however, is a system which deals with hierarchical data. In this paper, we present a comparison between the two previously mentioned protein descriptors for predicting protein function. Firstly, protein descriptors are extracted from the structural coordinates found in the Protein Data Bank (PDB) and proteins backbone, appropriately. Afterwards, GO class hierarchy is added to each protein which has descriptor data. This created file is used as an input to the CLUS system. CLUS generates a decision tree model which is trained from the protein structure. The results from this system are the GO classes in which the protein belongs. The generated output shows that the predicting protein function with the Voxel protein descriptor gives better results instead of predicting protein function with the Ray protein descriptor.

Study question What are the molecular pathways overactivated in arrested cleavage stage human embryos? Summary answer There is an upregulation of mitochondrial activity and cellular stress-related pathways in arrested cleavage embryos, which is in agreement with the “quiet hypothesis”. What is known already mtDNA content decreases during embryo development, however there is a high increase in mtDNA content in arrested cleavage embryos that may correspond to a response to intrinsic or extrinsic factor creating stress. This reasoning would be in agreement with stablished hypothesis showing that a basal metabolism with a moderate-low energy consumption is actually a sign of embryo health, so a viable embryo does not need to use an extra energy to accommodate to the environment (Leese, 2012). The study of the transcriptional profile during human embryo development will give further information about key molecular process involved in in-vitro embryonic competence. Study design, size, duration A prospective cohort study was performed with 11 MII oocytes (average age= 22,9 years), 10 non-arrested cleavage embryos (average age= 29,9 years), 5 arrested cleavage embryos (average age= 38,8 years) and 8 blastocysts (average age= 39,1 years). All specimens were warmed and sampled in PCR tubes with 2 μl of suitable reaction buffer for the RNA sequencing protocol. Participants/materials, setting, methods Specimens were analyzed by single-cell RNA sequencing (scRNA-seq). Correlation studies, principal component and differential expression analysis were performed with DESeq2 package. Differential gene expression analyses were done using the parametric Wald test, with Benjamini-Hochberg multiple test correction (padj). Finally, Fgsea algorithm was used for enrichment analysis on Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways and Gene ontology (GO) terms. Main results and the role of chance We do not observe mitochondria-related activity pathways significantly (P>.05) deregulated between MII oocytes and non-arrested cleavage embryos considering GO and KEGG categories. When comparing non-arrested cleavage embryos versus blastocysts, we observe several ATP production/consumption and cristae formation-related pathways significantly (P<.05) upregulated in blastocysts compared to non-arrested cleavage embryos considering GO and KEGG categories. This change in activity coincides with the metabolic activation event that occurs in the blastocyst stage. However, when we analyze arrested cleavage embryos versus non-arrested cleavage embryos, we observe several ATP production related-pathways and mitochondria-related apoptosis pathways significantly (P<.05) upregulated in arrested cleavage embryos compared to non-arrested cleavage embryos considering GO categories. With KEGG categories, we notice a significant (P<.05) upregulation of oxidative phosphorylation in arrested cleavage embryos. On the other hand, when we analyze the differences between arrested cleavage embryos and blastocysts, taking into account the differences related to the change of embryo stage, we do not observe ATP production or consumption-related pathways significantly (P>.05) deregulated considering GO and KEGG categories. Then, human arrested embryos in parallel with the increase in mtDNA content, display an upregulation of mitochondrial activity and cellular stress which is in line with the expected overactive metabolism of non-viable embryos. Limitations, reasons for caution All analyzed blastocyst were aneuploid, so we are unable to determine what the results would be with euploid blastocysts. Also, although age can it be ruled out, no differences were observed between mean age from cleavage stage embryos (arrested and non-arrested ones) and blastocyst. Wider implications of the findings: This study evidences the relation between extreme mtDNA content increase and the identification of the pathways involved in active metabolism and apoptosis in arrested cleavage stage human embryos. Trial registration number Not applicable

Approximately 9334 (37%) of human genes have no publications documenting their function and, for those that are published, the number of publications per gene is highly skewed. Furthermore, for reasons not clear, the entry of new gene names into the literature has slowed in recent years. If we are to better understand human/mammalian biology and complete the catalog of human gene function, it is important to finish predicting putative functions for these genes based upon existing experimental evidence. A global meta-analysis (GMA) of all publicly available GEO two-channel human microarray datasets (3551 experiments total) was conducted to identify genes with recurrent, reproducible patterns of co-regulation across different conditions. Patterns of co-expression were divided into parallel (i.e. genes are up and down-regulated together) and anti-parallel. Several ranking methods to predict a gene's function based on its top 20 co-expressed gene pairs were compared. In the best method, 34% of predicted Gene Ontology (GO) categories matched exactly with the known GO categories for approximately 5000 genes analyzed versus only 3% for random gene sets. Only 2.4% of co-expressed gene pairs were found as co-occurring gene pairs in MEDLINE. Via a GO enrichment analysis, genes co-expressed in parallel with the query gene were frequently associated with the same GO categories, whereas anti-parallel genes were not. Combining parallel and anti-parallel genes for analysis resulted in fewer significant GO categories, suggesting they are best analyzed separately. Expression databases contain much unexpected genetic knowledge that has not yet been reported in the literature. A total of 1642 Human genes with unknown function were differentially expressed in at least 30 experiments. Data matrix available upon request.

Motivation: The automated functional annotation of biological macromolecules is a problem of computational assignment of biological concepts or ontological terms to genes and gene products. A number of methods have been developed to computationally annotate genes using standardized nomenclature such as Gene Ontology (GO). However, questions remain about the possibility for development of accurate methods that can integrate disparate molecular data as well as about an unbiased evaluation of these methods. One important concern is that experimental annotations of proteins are incomplete. This raises questions as to whether and to what degree currently available data can be reliably used to train computational models and estimate their performance accuracy.Results: We study the effect of incomplete experimental annotations on the reliability of performance evaluation in protein function prediction. Using the structured-output learning framework, we provide theoretical analyses and carry out simulations to characterize the effect of growing experimental annotations on the correctness and stability of performance estimates corresponding to different types of methods. We then analyze real biological data by simulating the prediction, evaluation and subsequent re-evaluation (after additional experimental annotations become available) of GO term predictions. Our results agree with previous observations that incomplete and accumulating experimental annotations have the potential to significantly impact accuracy assessments. We find that their influence reflects a complex interplay between the prediction algorithm, performance metric and underlying ontology. However, using the available experimental data and under realistic assumptions, our results also suggest that current large-scale evaluations are meaningful and almost surprisingly reliable.Contact:predrag@indiana.eduSupplementary information:Supplementary data are available at Bioinformatics online.

Mild traumatic brain injuries (mTBIs) are one of the most prevalent neurological disorders, and humans are severely limited in their ability to repair and regenerate central nervous system (CNS) tissue postinjury. However, zebrafish (Danio rerio) maintain the remarkable ability to undergo complete and functional neuroregeneration as an adult. We wish to extend knowledge of the known mechanisms of neuroregeneration by analyzing the differentially expressed genes (DEGs) in a novel adult zebrafish model of mTBI. In this study, a rodent weight drop model of mTBI was adapted to the adult zebrafish. A memory test showed significant deficits in spatial memory in the mTBI group. We identified DEGs at 3 and 21 days postinjury (dpi) through RNA-sequencing analysis. The resulting DEGs were categorized according to gene ontology (GO) categories. At 3 dpi, GO categories consisted of peak injury response pathways. Significantly, at 21 dpi, GO categories consisted of neuroregeneration pathways. Ultimately, these results validate a novel zebrafish model of mTBI and elucidate significant DEGs of interest in CNS injury and neuroregeneration.

Advances in mass spectrometry-based proteomics have yielded a substantial mapping of the tyrosine phosphoproteome and thus provided an important step toward a systematic analysis of intracellular signaling networks in higher eukaryotes. In this study we decomposed an uncharacterized proteomics data set of 481 unique phosphotyrosine (Tyr(P)) peptides by sequence similarity to known ligands of the Src homology 2 (SH2) and the phosphotyrosine binding (PTB) domains. From 20 clusters we extracted 16 known and four new interaction motifs. Using quantitative mass spectrometry we pulled down Tyr(P)-specific binding partners for peptides corresponding to the extracted motifs. We confirmed numerous previously known interaction motifs and found 15 new interactions mediated by phosphosites not previously known to bind SH2 or PTB. Remarkably, a novel hydrophobic N-terminal motif ((L/V/I)(L/V/I)pY) was identified and validated as a binding motif for the SH2 domain-containing inositol phosphatase SHIP2. Our decomposition of the in vivo Tyr(P) proteome furthermore suggests that two-thirds of the Tyr(P) sites mediate interaction, whereas the remaining third govern processes such as enzyme activation and nucleic acid binding.

Non-negative matrix factorization and differential expression analyses identify hub genes linked to progression and prognosis of glioblastoma multiforme

BackgroundIn the current climate of high-throughput computational biology, the inference of a protein's function from related measurements, such as protein-protein interaction relations, has become a canonical task. Most existing technologies pursue this task as a classification problem, on a term-by-term basis, for each term in a database, such as the Gene Ontology (GO) database, a popular rigorous vocabulary for biological functions. However, ontology structures are essentially hierarchies, with certain top to bottom annotation rules which protein function predictions should in principle follow. Currently, the most common approach to imposing these hierarchical constraints on network-based classifiers is through the use of transitive closure to predictions.ResultsWe propose a probabilistic framework to integrate information in relational data, in the form of a protein-protein interaction network, and a hierarchically structured database of terms, in the form of the GO database, for the purpose of protein function prediction. At the heart of our framework is a factorization of local neighborhood information in the protein-protein interaction network across successive ancestral terms in the GO hierarchy. We introduce a classifier within this framework, with computationally efficient implementation, that produces GO-term predictions that naturally obey a hierarchical 'true-path' consistency from root to leaves, without the need for further post-processing.ConclusionA cross-validation study, using data from the yeast Saccharomyces cerevisiae, shows our method offers substantial improvements over both standard 'guilt-by-association' (i.e., Nearest-Neighbor) and more refined Markov random field methods, whether in their original form or when post-processed to artificially impose 'true-path' consistency. Further analysis of the results indicates that these improvements are associated with increased predictive capabilities (i.e., increased positive predictive value), and that this increase is consistent uniformly with GO-term depth. Additional in silico validation on a collection of new annotations recently added to GO confirms the advantages suggested by the cross-validation study. Taken as a whole, our results show that a hierarchical approach to network-based protein function prediction, that exploits the ontological structure of protein annotation databases in a principled manner, can offer substantial advantages over the successive application of 'flat' network-based methods.

Protein function prediction is a crucial task in bioinformatics and computational biology, as it enables the understanding of disease mechanisms, development of new therapeutics, and improvement of crop yields. Despite significant advances, the majority of protein functions remain unknown or poorly annotated, hindering our understanding of biological systems. This review provides a comprehensive overview of the available methods for protein function prediction, categorizing them into eight classes based on the sources of information they use. We examine over 35 methods, including traditional sequence-based approaches and recent advances in machine learning and natural language processing. We also discuss the incorporation of background knowledge in Gene Ontology and zero-shot predictions. To improve protein function prediction, we highlight the need for developing more accurate and robust methods that integrate multiple sources of information. We provide several practical notes for choosing and interpreting the results of protein function prediction methods.

BackgroundThe Gene Ontology (GO) Consortium organizes genes into hierarchical categories based on biological process, molecular function and subcellular localization. Tools such as GoMiner can leverage GO to perform ontological analysis of microarray and proteomics studies, typically generating a list of significant functional categories. Two or more of the categories are often redundant, in the sense that identical or nearly-identical sets of genes map to the categories. The redundancy might typically inflate the report of significant categories by a factor of three-fold, create an illusion of an overly long list of significant categories, and obscure the relevant biological interpretation.ResultsWe now introduce a new resource, RedundancyMiner, that de-replicates the redundant and nearly-redundant GO categories that had been determined by first running GoMiner. The main algorithm of RedundancyMiner, MultiClust, performs a novel form of cluster analysis in which a GO category might belong to several category clusters. Each category cluster follows a "complete linkage" paradigm. The metric is a similarity measure that captures the overlap in gene mapping between pairs of categories.ConclusionsRedundancyMiner effectively eliminated redundancies from a set of GO categories. For illustration, we have applied it to the clarification of the results arising from two current studies: (1) assessment of the gene expression profiles obtained by laser capture microdissection (LCM) of serial cryosections of the retina at the site of final optic fissure closure in the mouse embryos at specific embryonic stages, and (2) analysis of a conceptual data set obtained by examining a list of genes deemed to be "kinetochore" genes.

Molecular fingerprinting of the podocyte reveals novel gene and protein regulatory networks

Drought is a significant constraint to yield increase in soybean. The early perception of water deprivation is critical for recruitment of genes that promote plant tolerance. DeepSuperSAGE libraries, including one control and a bulk of six stress times imposed (from 25 to 150 min of root dehydration) for drought-tolerant and sensitive soybean accessions, allowed to identify new molecular targets for drought tolerance. The survey uncovered 120,770 unique transcripts expressed by the contrasting accessions. Of these, 57,610 aligned with known cDNA sequences, allowing the annotation of 32,373 unitags. A total of 1,127 unitags were up-regulated only in the tolerant accession, whereas 1,557 were up-regulated in both as compared to their controls. An expression profile concerning the most representative Gene Ontology (GO) categories for the tolerant accession revealed the expression “protein binding” as the most represented for “Molecular Function”, whereas CDPK and CBL were the most up-regulated protein families in this category. Furthermore, particular genes expressed different isoforms according to the accession, showing the potential to operate in the distinction of physiological behaviors. Besides, heat maps comprising GO categories related to abiotic stress response and the unitags regulation observed in the expression contrasts covering tolerant and sensitive accessions, revealed the unitags potential for plant breeding. Candidate genes related to “hormone response” (LOX, ERF1b, XET), “water response” (PUB, BMY), “salt stress response” (WRKY, MYB) and “oxidative stress response” (PER) figured among the most promising molecular targets. Additionally, nine transcripts (HMGR, XET, WRKY20, RAP2-4, EREBP, NAC3, PER, GPX5 and BMY) validated by RT-qPCR (four different time points) confirmed their differential expression and pointed that already after 25 minutes a transcriptional reorganization started in response to the new condition, with important differences between both accessions.

While genetic influences on alcohol dependence (AD) are substantial, progress in the identification of individual genetic variants that impact on risk has been difficult. We performed a genome-wide association study on 3,169 alcohol consuming subjects from the population-based Molecular Genetics of Schizophrenia (MGS2) control sample. Subjects were asked 7 questions about symptoms of AD which were analyzed by confirmatory factor analysis. Genotyping was performed using the Affymetrix 6.0 array. Three sets of analyses were conducted separately for European American (EA, n = 2,357) and African-American (AA, n = 812) subjects: individual single nucleotide polymorphisms (SNPs), candidate genes and enriched pathways using gene ontology (GO) categories. The symptoms of AD formed a highly coherent single factor. No SNP approached genome-wide significance. In the EA sample, the most significant intragenic SNP was in KCNMA1, the human homolog of the slo-1 gene in C. Elegans. Genes with clusters of significant SNPs included AKAP9, phosphatidylinositol glycan anchor biosynthesis, class G (PIGG), and KCNMA1. In the AA sample, the most significant intragenic SNP was CEACAM6 and genes showing empirically significant SNPs included KCNQ5, SLC35B4, and MGLL. In the candidate gene based analyses, the most significant findings were with ADH1C, nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 (NFKB1) and ankyrin repeat and kinase domain containing 1 (ANKK1) in the EA sample, and ADH5, POMC, and CHRM2 in the AA sample. The ALIGATOR program identified a significant excess of associated SNPs within and near genes in a substantial number of GO categories over a range of statistical stringencies in both the EA and AA sample. While we cannot be highly confident about any single result from these analyses, a number of findings were suggestive and worthy of follow-up. Although quite large samples will be needed to obtain requisite power, the study of AD symptoms in general population samples is a viable complement to case-control studies in identifying genetic risk variants for AD.

BackgroundGrapevine is a major food crop that is affected by global climate change. Consistent with field studies, dehydration assays of grapevine leaves can reveal valuable information of the plant’s response at physiological, transcript, and protein levels. There are well-known differences in grapevine rootstocks responses to dehydration. We used time-series transcriptomic approaches combined with network analyses to elucidate and identify important physiological processes and network hubs that responded to dehydration in three different grapevine species differing in their drought tolerance.ResultsTranscriptomic analyses of the leaves of Cabernet Sauvignon, Riparia Gloire, and Ramsey were evaluated at different times during a 24-h controlled dehydration. Analysis of variance (ANOVA) revealed that approximately 11,000 transcripts changed significantly with respect to the genotype x treatment interaction term and approximately 6000 transcripts changed significantly according to the genotype x treatment x time interaction term indicating massive differential changes in gene expression over time. Standard analyses determined substantial effects on the transcript abundance of genes involved in the metabolism and signaling of two known plant stress hormones, abscisic acid (ABA) and ethylene. ABA and ethylene signaling maps were constructed and revealed specific changes in transcript abundance that were associated with the known drought tolerance of the genotypes including genes such as VviABI5, VviABF2, VviACS2, and VviWRKY22. Weighted-gene coexpression network analysis (WGCNA) confirmed these results. In particular, WGCNA identified 30 different modules, some of which had highly enriched gene ontology (GO) categories for photosynthesis, phenylpropanoid metabolism, ABA and ethylene signaling. The ABA signaling transcription factors, VviABI5 and VviABF2, were highly connected hubs in two modules, one being enriched in gaseous transport and the other in ethylene signaling. VviABI5 was distinctly correlated with an early response and high expression for the drought tolerant Ramsey and with little response from the drought sensitive Riparia Gloire. These ABA signaling transcription factors were highly connected to VviSnRK1 and other gene hubs associated with sugar, ethylene and ABA signaling.ConclusionA leaf dehydration assay provided transcriptomic evidence for differential leaf responses to dehydration between genotypes differing in their drought tolerance. WGCNA proved to be a powerful network analysis approach; it identified 30 distinct modules (networks) with highly enriched GO categories and enabled the identification of gene hubs in these modules. Some of these genes were highly connected hubs in both the ABA and ethylene signaling pathways, supporting the hypothesis that there is substantial crosstalk between the two hormone pathways. This study identifies solid gene candidates for future investigations of drought tolerance in grapevine.Electronic supplementary materialThe online version of this article (doi:10.1186/s12870-016-0804-6) contains supplementary material, which is available to authorized users.

Transcriptome analysis is important for interpreting the functional elements of the genome and revealing the molecular constituents of cells and tissues. Herein, differentially transcribed genes were identified by deep sequencing after zebrafish (Danio rerio) were exposed to β-diketone antibiotics (DKAs); 23,129 and 23,550 mapped genes were detected in control and treatment groups, a total of 3238 genes were differentially expressed between control and treatment groups. Of these genes, 328 genes (213 up- and 115 down-regulation) had significant differential expression (p < 0.05) and an expression ratio (control/treatment) of >2 or <0.5. Additionally, we performed Gene Ontology (GO) category and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, and found 266 genes in the treatment group with annotation terms linked to the GO category. A total of 77 differentially expressed transcriptional genes were associated with 132 predicted KEGG metabolic pathways. Serious liver tissue damage was reflected and consistent with the differences in genetic classification and function from the transcriptome analysis. These results enhance our understanding of zebrafish developmental processes under exposure to DKA stress. © 2015 Wiley Periodicals, Inc. Environ Toxicol 31: 1357-1371, 2016.