Automated protein function prediction is a challenging problem with distinctive features, such as the hierarchical organization of protein functions and the scarcity of annotated proteins for most biological functions. We propose a multitask learning algorithm addressing both issues. Unlike standard multitask algorithms, which use task (protein functions) similarity information as a bias to speed up learning, we show that dissimilarity information enforces separation of rare class labels from frequent class labels, and for this reason is better suited for solving unbalanced protein function prediction problems. We support our claim by showing that a multitask extension of the label propagation algorithm empirically works best when the task relatedness information is represented using a dissimilarity matrix as opposed to a similarity matrix. Moreover, the experimental comparison carried out on three model organism shows that our method has a more stable performance in both "protein-centric" and "function-centric" evaluation settings.
Multitask Protein Function Prediction Through Task Dissimilarity / M. Frasca, N..A. Cesa Bianchi. - In: IEEE/ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS. - ISSN 1545-5963. - 16:5(2019), pp. 1550-1560. [10.1109/TCBB.2017.2684127]
Multitask Protein Function Prediction Through Task Dissimilarity
M. Frasca
;N..A. Cesa Bianchi
2019
Abstract
Automated protein function prediction is a challenging problem with distinctive features, such as the hierarchical organization of protein functions and the scarcity of annotated proteins for most biological functions. We propose a multitask learning algorithm addressing both issues. Unlike standard multitask algorithms, which use task (protein functions) similarity information as a bias to speed up learning, we show that dissimilarity information enforces separation of rare class labels from frequent class labels, and for this reason is better suited for solving unbalanced protein function prediction problems. We support our claim by showing that a multitask extension of the label propagation algorithm empirically works best when the task relatedness information is represented using a dissimilarity matrix as opposed to a similarity matrix. Moreover, the experimental comparison carried out on three model organism shows that our method has a more stable performance in both "protein-centric" and "function-centric" evaluation settings.File | Dimensione | Formato | |
---|---|---|---|
MTLPFrascaCesaBianchi.pdf
accesso aperto
Descrizione: Articolo principale
Tipologia:
Pre-print (manoscritto inviato all'editore)
Dimensione
328.71 kB
Formato
Adobe PDF
|
328.71 kB | Adobe PDF | Visualizza/Apri |
07880576.pdf
accesso riservato
Tipologia:
Publisher's version/PDF
Dimensione
539.6 kB
Formato
Adobe PDF
|
539.6 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.