Using the Gene Ontology Hierarchy when Predicting Gene Function

Sara Mostafavi, Quaid Morris
Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, PMLR R7:427-435, 2009.

Abstract

The problem of multilabel classification when the labels are related through a hierarchical categorization scheme occurs in many application domains such as computational biology. For example, this problem arises naturally when trying to automatically assign gene function using a controlled vocabularies like Gene Ontology. However, most existing approaches for predicting gene functions solve independent classification problems to predict genes that are involved in a given function category, independently of the rest. Here, we propose two simple methods for incorporating information about the hierarchical nature of the categorization scheme. In the first method, we use information about a gene’s previous annotation to set an initial prior on its label. In a second approach, we extend a graph-based semi-supervised learning algorithm for predicting gene function in a hierarchy. We show that we can efficiently solve this problem by solving a linear system of equations. We compare these approaches with a previous label reconciliation-based approach. Results show that using the hierarchy information directly, compared to using reconciliation methods, improves gene function prediction.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR7-mostafavi09a, title = {Using the Gene Ontology Hierarchy when Predicting Gene Function}, author = {Mostafavi, Sara and Morris, Quaid}, booktitle = {Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence}, pages = {427--435}, year = {2009}, editor = {Bilmes, Jeff and Ng, Andrew Y.}, volume = {R7}, series = {Proceedings of Machine Learning Research}, month = {18--21 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r7/main/assets/mostafavi09a/mostafavi09a.pdf}, url = {https://proceedings.mlr.press/r7/mostafavi09a.html}, abstract = {The problem of multilabel classification when the labels are related through a hierarchical categorization scheme occurs in many application domains such as computational biology. For example, this problem arises naturally when trying to automatically assign gene function using a controlled vocabularies like Gene Ontology. However, most existing approaches for predicting gene functions solve independent classification problems to predict genes that are involved in a given function category, independently of the rest. Here, we propose two simple methods for incorporating information about the hierarchical nature of the categorization scheme. In the first method, we use information about a gene’s previous annotation to set an initial prior on its label. In a second approach, we extend a graph-based semi-supervised learning algorithm for predicting gene function in a hierarchy. We show that we can efficiently solve this problem by solving a linear system of equations. We compare these approaches with a previous label reconciliation-based approach. Results show that using the hierarchy information directly, compared to using reconciliation methods, improves gene function prediction.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Using the Gene Ontology Hierarchy when Predicting Gene Function %A Sara Mostafavi %A Quaid Morris %B Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2009 %E Jeff Bilmes %E Andrew Y. Ng %F pmlr-vR7-mostafavi09a %I PMLR %P 427--435 %U https://proceedings.mlr.press/r7/mostafavi09a.html %V R7 %X The problem of multilabel classification when the labels are related through a hierarchical categorization scheme occurs in many application domains such as computational biology. For example, this problem arises naturally when trying to automatically assign gene function using a controlled vocabularies like Gene Ontology. However, most existing approaches for predicting gene functions solve independent classification problems to predict genes that are involved in a given function category, independently of the rest. Here, we propose two simple methods for incorporating information about the hierarchical nature of the categorization scheme. In the first method, we use information about a gene’s previous annotation to set an initial prior on its label. In a second approach, we extend a graph-based semi-supervised learning algorithm for predicting gene function in a hierarchy. We show that we can efficiently solve this problem by solving a linear system of equations. We compare these approaches with a previous label reconciliation-based approach. Results show that using the hierarchy information directly, compared to using reconciliation methods, improves gene function prediction. %Z Reissued by PMLR on 04 October 2026.
APA
Mostafavi, S. & Morris, Q.. (2009). Using the Gene Ontology Hierarchy when Predicting Gene Function. Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R7:427-435 Available from https://proceedings.mlr.press/r7/mostafavi09a.html. Reissued by PMLR on 04 October 2026.

Related Material