Integrative approaches to the prediction of protein functions based on the feature selection

PhyloGene server for identification and visualization of co-evolving proteins using normalized phylogenetic profiles

Proteins that function in the same pathways, protein complexes or the same environmental conditions can show similar patterns of sequence conservation across phylogenetic clades. In species that no longer require a specific protein complex or pathway, these proteins, as a group, tend to be lost or diverge. Analysis of the similarity in patterns of sequence conservation across a large set of eukaryotes can predict functional associations between different proteins, identify new pathway members and reveal the function of previously uncharacterized proteins. We used normalized phylogenetic profiling to predict protein function and identify new pathway members and disease genes. The phylogenetic profiles of tens of thousands conserved proteins in the human, mouse, Caenorhabditis elegans and Drosophila genomes can be queried on our new web server, PhyloGene. PhyloGene provides intuitive and user-friendly platform to query the patterns of conservation across 86 animal, fungal, plant and protist genomes. A protein query can be submitted either by selecting the name from whole-genome protein sets of the intensively studied species or by entering a protein sequence. The graphic output shows the profile of sequence conservation for the query and the most similar phylogenetic profiles for the proteins in the genome of choice. The user can also download this output in numerical form.

Gene function prediction based on the Gene Ontology hierarchical structure

The information of the Gene Ontology annotation is helpful in the explanation of life science phenomena, and can provide great support for the research of the biomedical field. The use of the Gene Ontology is gradually affecting the way people store and understand bioinformatic data. To facilitate the prediction of gene functions with the aid of text mining methods and existing resources, we transform it into a multi-label top-down classification problem and develop a method that uses the hierarchical relationships in the Gene Ontology structure to relieve the quantitative imbalance of positive and negative training samples. Meanwhile the method enhances the discriminating ability of classifiers by retaining and highlighting the key training samples. Additionally, the top-down classifier based on a tree structure takes the relationship of target classes into consideration and thus solves the incompatibility between the classification results and the Gene Ontology structure. Our experiment on the Gene Ontology annotation corpus achieves an F-value performance of 50.7% (precision: 52.7% recall: 48.9%). The experimental results demonstrate that when the size of training set is small, it can be expanded via topological propagation of associated documents between the parent and child nodes in the tree structure. The top-down classification model applies to the set of texts in an ontology structure or with a hierarchical relationship.

Predicting drug-target interactions using drug-drug interactions

Computational methods for predicting drug-target interactions have become important in drug research because they can help to reduce the time, cost, and failure rates for developing new drugs. Recently, with the accumulation of drug-related data sets related to drug side effects and pharmacological data, it has became possible to predict potential drug-target interactions. In this study, we focus on drug-drug interactions (DDI), their adverse effects ([Formula: see text]) and pharmacological information ([Formula: see text]), and investigate the relationship among chemical structures, side effects, and DDIs from several data sources. In this study, [Formula: see text] data from the STITCH database, [Formula: see text] from drugs.com, and drug-target pairs from ChEMBL and SIDER were first collected. Then, by applying two machine learning approaches, a support vector machine (SVM) and a kernel-based L1-norm regularized logistic regression (KL1LR), we showed that DDI is a promising feature in predicting drug-target interactions. Next, the accuracies of predicting drug-target interactions using DDI were compared to those obtained using the chemical structure and side effects based on the SVM and KL1LR approaches, showing that DDI was the data source contributing the most for predicting drug-target interactions.

Human disease locus discovery and mapping to molecular pathways through phylogenetic profiling

By analyzing the conservation of human proteins across 87 species, we sorted proteins into clusters of coevolution. Some clusters are enriched for genes assigned to particular human diseases or molecular pathways; the other genes in the same cluster may function in related pathways and diseases.

Many genes that were thought to map to different diseases are actually coevolved together and mapped into the same phylogenetic clusters.

Many molecular pathways map to the same phylogenetic clusters as genes associated with specific human diseases.

Focusing on proteins coevolved with the microphthalmia-associated transcription factor (MITF), we identified the Notch pathway suppressor of hairless (RBP-Jk/SuH) transcription factor, and showed that RBP-Jk functions as an MITF cofactor.

Our analysis thus establishes a connectivity between different diseases and pathways, linking diseases phenotypes and functional gene groups.

Genes with common profiles of the presence and absence in disparate genomes tend to function in the same pathway. By mapping all human genes into about 1000 clusters of genes with similar patterns of conservation across eukaryotic phylogeny, we determined that sets of genes associated with particular diseases have similar phylogenetic profiles. By focusing on those human phylogenetic gene clusters that significantly overlap some of the thousands of human gene sets defined by their coexpression or annotation to pathways or other molecular attributes, we reveal the evolutionary map that connects molecular pathways and human diseases. The other genes in the phylogenetic clusters enriched for particular known disease genes or molecular pathways identify candidate genes for roles in those same disorders and pathways. Focusing on proteins coevolved with the microphthalmia-associated transcription factor (MITF), we identified the Notch pathway suppressor of hairless (RBP-Jk/SuH) transcription factor, and showed that RBP-Jk functions as an MITF cofactor.

A Resource of Quantitative Functional Annotation for Homo sapiens Genes

The body of human genomic and proteomic evidence continues to grow at ever-increasing rates, while annotation efforts struggle to keep pace. A surprisingly small fraction of human genes have clear, documented associations with specific functions, and new functions continue to be found for characterized genes. Here we assembled an integrated collection of diverse genomic and proteomic data for 21,341 human genes and make quantitative associations of each to 4333 Gene Ontology terms. We combined guilt-by-profiling and guilt-by-association approaches to exploit features unique to the data types. Performance was evaluated by cross-validation, prospective validation, and by manual evaluation with the biological literature. Functional-linkage networks were also constructed, and their utility was demonstrated by identifying candidate genes related to a glioma FLN using a seed network from genome-wide association studies. Our annotations are presented—alongside existing validated annotations—in a publicly accessible and searchable web interface.

Fusion methodologies for biomedical data

Data fusion methods are powerful tools for integrating the different views of an organism provided by various types of experimental data. We describe various methodologies for integrating and drawing inferences from a collection of biomedical data, primarily focusing on protein and gene expression data. Computational experiments performed using biomedical data, including known protein-protein interactions, hydropathy profiles, gene expression data and amino acid sequences, demonstrate the utility of this approach. Overall, studies agree in that methodologies using carefully selected data of various types to predict particular classes, groups and interactions, perform better than when applied to a single type of data.