Advanced search
Start date
Betweenand


Prediction of Protein Molecular Functions Using Transformers

Full text
Author(s):
de Mello, Felipe Lopes ; de Oliveira, Gabriel Bianchin ; Pedrini, Helio ; Dias, Zanoni ; Rutkowski, L ; Scherer, R ; Korytkowski, M ; Pedrycz, W ; Tadeusiewicz, R ; Zurada, JM
Total Authors: 10
Document type: Journal article
Source: ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING, ICAISC 2022, PT II; v. 13589, p. 9-pg., 2023-01-01.
Abstract

At the end of 2021, there were more than 200 million proteins in which their molecular functions were still unknown. As the empirical determination of these functions is slow and expensive, several research groups around the world have applied machine learning to perform the prediction of protein functions. In this work, we evaluate the use of Transformer architectures to classify protein molecular functions. Our classifier uses the embeddings resulting from two Transformer-based architectures as input to a Multi-Layer Perceptron classifier. This model got Fmax of 0.562 in our database and, when we applied this model to the same database used by DeepGOPlus, we reached the value of 0.617, surpassing the best result available in the literature. (AU)

FAPESP's process: 17/16246-0 - Sensitive media analysis through deep learning architectures
Grantee:Sandra Eliza Fontes de Avila
Support Opportunities: Regular Research Grants
FAPESP's process: 19/20875-8 - Chest X-ray image classification using deep neural networks
Grantee:Vinicius Teixeira de Melo
Support Opportunities: Scholarships in Brazil - Master
FAPESP's process: 17/12646-3 - Déjà vu: feature-space-time coherence from heterogeneous data for media integrity analytics and interpretation of events
Grantee:Anderson de Rezende Rocha
Support Opportunities: Research Projects - Thematic Grants
FAPESP's process: 15/11937-9 - Investigation of hard problems from the algorithmic and structural stand points
Grantee:Flávio Keidi Miyazawa
Support Opportunities: Research Projects - Thematic Grants