Predicting Gene Function From Patterns of Annotation

Abstract

The Gene Ontology (GO) Consortium has produced a controlled vocabulary for annotation of gene function that is used in many organism-specific gene annotation databases. This allows the prediction of gene function based on patterns of annotation. For example, if annotations for two attributes tend to occur together in a database, then a gene holding one attribute is likely to hold the other as well. We modeled the relationships among GO attributes with decision trees and Bayesian networks, using the annotations in the Saccharomyces Genome Database (SGD) and in FlyBase as training data. We tested the models using cross-validation, and we manually assessed 100 gene-attribute associations that were predicted by the models but that were not present in the SGD or FlyBase databases. Of the 100 manually assessed associations, 41 were judged to be true, and another 42 were judged to be plausible.[Detailed lists of hypotheses including the curators' comments on each hypothesis, are available at

Gene expression correlation and gene ontology-based similarity: an assessment of quantitative relationships

“…Machine learning models have been previously reported to address this problem. However, they measure similarity between sets of annotations based solely on the presence or absence of GO terms [9]. Thus, the information-theoretic tools evaluated in this paper may be useful to support the development of more meaningful and reliable prediction models.…”

“…They applied rough set theory to assign biological process terms to genes represented by expression patterns. King et al [9] implemented decision trees and Bayesian networks to predict new GO terms-gene associations based on existing annotations from the SGD and FlyBase. Laegreid et al [10] also applied supervised learning methods to predict GO biological process annotation terms.…”

Anomaly-free Prediction of Gene Ontology Annotations Using Bayesian Networks

“…A few years ago, King et al [3] proposed the use of decision trees and Bayesian networks for predicting annotations by learning patterns from available annotation profiles. Recently, Tao et al [4] proposed to use a k-nearest neighborg (k-NN) classifier, whereby a gene inherits the annotations that are common among its nearest neighbor genes, determined according to the functional distance between genes, based on the semantic similarity of GO terms used to annotate them.…”

Instance-Based Matching of Large Life Science Ontologies

“…Similar to our study, it applies the Kappa similarity metric including a significance test. [9] applies decision trees and Bayesian networks to create matches between GO subontologies that is different to our approach. It uses available annotations (instances) of two species (mouse and human) as training data and for cross validation to test the models.…”

“…Instance-based ontology matching is investigated in [8,9,3,11]. They follow statistical or machine learning approaches and apply them in different application domains.…”