Large-Scale Protein Annotation through Gene Ontology

Abstract

Recent progress in genomic sequencing, computational biology, and ontology development has presented an opportunity to investigate biological systems from a unique perspective, that is, examining genomes and transcriptomes through the multiple and hierarchical structure of Gene Ontology (GO). We report here our development of GO Engine, a computational platform for GO annotation, and analysis of the resultant GO annotations of human proteins. Protein annotation was centered on sequence homology with GO-annotated proteins and protein domain analysis. Text information analysis and a multiparameter cellular localization predictive tool were also used to increase the annotation accuracy, and to predict novel annotations. The majority of proteins corresponding to full-length mRNA in GenBank, and the majority of proteins in the NR database (nonredundant database of proteins) were annotated with one or more GO nodes in each of the three GO categories. The annotations of GenBank and SWISS-PROT proteins are available to the public at the GO Consortium web site.Biomedical research over the last century has made tremendous progress in our understanding of biology and medicine. The recent genomic sequencing of human, mouse, and other organisms, and high-throughput studies, such as those based on microarray technology, have been yielding massive amounts of data. However, the knowledge accumulated so far is mainly fragmented. Full utilization of this data and its integration with existing knowledge can be facilitated by a systematic representation of knowledge, that is, the development of ontology. Ontology is the formalized specification of knowledge in a certain subject. Great potential exists for ontology-based literature retrieval in biomedical research (McGuinness 1999), ontology-based database integration in drug discovery, and ontology-facilitated biomedical research. Recently, the Gene Ontology (GO) Consortium (www. geneontology.org) has developed a systematic and standardized nomenclature for annotating genes in various organisms. Using three main ontologies-molecular function, biological process, and cellular component-a significant number of genes in yeast, Drosophila, mouse, and other model organisms have been annotated, either manually or automatically (Ashburner et al. 2000; The Gene Ontology Consortium 2001).Association between ontology nodes and proteins, namely, protein annotation through gene ontology, is an integral application of ontology and has many practical uses. For example, designing of microarray probes would be greatly facilitated by a comprehensive understanding of all the genes involved. A microarray aimed to examine a particular process, such as apoptosis, would optimally have probes against all the genes significantly and directly involved in apoptosis. These genes can be chosen using GO annotations.To efficiently annotate proteins, we have developed a software platform, the GO Engine, which combines rigorous sequence homology comparison with text information analysis. During evolution, many ne...

Protein Profiling of Mouse Livers with Peroxisome Proliferator-Activated Receptor α Activation

“…Nine of these 12 proteins were up-regulated in both AOX Ϫ/Ϫ and Wy-14,643-treated mice in comparison to levels in wild-type mice. Since fatty acid metabolism has been studied to a great extent in peroxisome proliferation, 7 of the 12 proteins, AOX (4,25,44,51), mediumchain acyl-CoA dehydrogenase (7,34), acyl-CoA thioesterase 1 (9, 47, 48), 3-hydroxy-3-methylglutaryl-CoA synthase 1 (45), 3-ketoacyl-CoA thiolase A (33, 44), acyl-CoA thioester hydrolase (2), and L-peroxisomal bifunctional enzyme (33,37,44), had previously been reported to be regulated in AOX Ϫ/Ϫ mice or regulated by peroxisome proliferator treatment due to PPAR␣ activation. These gene products contain functional peroxisome proliferator response elements (PPRE) in the promoter region (8,40,41).…”

“…We report here the analysis of differential protein expression in AOX Ϫ/Ϫ and Wy-14,643-treated wild-type mice and the identification of 46 differentially expressed proteins. By gene ontology annotation (1,51), these differentially expressed proteins were aligned according to their primary function. These observations establish the similarities in protein profiles in livers with PPAR␣ activation by natural and synthetic ligands.…”

Combining Multisource Information Through Functional-Annotation-Based Weighting: Gene Function Prediction in Yeast

“…As compared to direct protein similarity search, the field of searching gene/protein similarity through phylogenetic profiles (PPs) [7], RS sequence [8], and transitive homology [18] are relatively new methods. In this investigation, transitive homologues are used instead of PP and RS for extracting protein similarity, as its performance is reported to be better than PP and RS in literature [19], [20].…”

An Empirical Evaluation of the Effectiveness of Different Types of Predictor Attributes in Protein Function Prediction

“…sequence length, presence/absence of a particular motif) and the classes correspond to the different functions that the protein can perform. For instance, in [13] neural networks were trained using sequence derived features such as amino acid biochemical properties and secondary structure; [15] used protein interaction data to create a probabilistic model based on markov random fields; PROSITE patterns were used in [19] to extract classification rules using C4.5; and [29] followed a clustering approach using protein domains and textual information from MEDLINE. For further examples refer to [23], [5] and [9].…”