Investigation of Multi-task Deep Neural Networks in Automated Protein Function Prediction

Date

2017-07-20

Author

Rifaioğlu, Ahmet Süreyya
Martin, Maria Jesus
Atalay, Rengül
Atalay, Mehmet Volkan
Doğan, Tunca

Metadata

Show full item record

Item Usage Stats

238
views

0
downloads

Functional annotation of proteins is a crucial research field for understanding molecular mechanisms of living-beings and for biomedical purposes (e.g. identification of disease-causing functional changes in genes and for discovering novel drugs). Several Gene Ontology (GO) based protein function prediction methods have been proposed in the last decade to annotate proteins. However, considering the prediction performances of the proposed methods, it can be stated that there is still room for significant improvements in protein function prediction area (1). Deep learning techniques became popular in recent years and turned out to be an industry standard in several areas such as computer vision and speech recognition. To the best of our knowledge, as of today, deep learning algorithms have not been applied to the large-scale protein function prediction problem. Here, we propose a hierarchical multi-task deep neural network architecture, DEEPred, as a solution to protein function prediction problem. First of all, we investigated the potential of employing deep learning methods for protein function prediction. For this purpose, we measured the performance of our models at different parameter settings. Furthermore, we examined the relationship between the performance of the system and the size of the training datasets, since the training set size has been reported in the literature to be significantly affecting the performance of deep learning models.

URI

https://www.iscb.org/ismbeccb2017
https://hdl.handle.net/11511/73348
https://www.biofunctionprediction.org/papers/Function_SIG_2017_paper_29.pdf

Conference Name

ISMB/ECCB 2017: 25th Annual International Conference on Intelligent Systems for Molecular Biology, (21 - 25 Temmuz 2017)

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Inference of Gene Regulatory Networks Via Multiple Data Sources and a Recommendation Method Ozsoy, Makbule Gulcin; Polat, Faruk; Alhajj, Reda (2015-11-12) Gene regulatory networks (GRNs) are composed of biological components, including genes, proteins and metabolites, and their interactions. In general, computational methods are used to infer the connections among these components. However, computational methods should take into account the general features of the GRNs, which are sparseness, scale-free topology, modularity and structure of the inferred networks. In this work, observing the common aspects between recommendation systems and GRNs, we decided to ...
Applications of the multifunctional magnetic nanoparticles for development of molecular therapies for breast cancer Aşık, Elif; Güray, Tülin; Volkan, Mürvet; Department of Biotechnology (2015) The understanding of how magnetic nanoparticles (MNPs) interact with living system is one of the prerequisite pieces of information needed to be obtained before any further development for desired biomedical applications. In this study, Cobalt Ferrite magnetic nanoparticles (CoFe-MNPs) in their naked and silica-coated forms were characterized. In vitro cell culture for their likely cytotoxicity and genotoxicity potential were examined. The apoptosis, lipid peroxidation, ROS formation and oxidative stress re...
Comparison of XL-MS Software to Identify Protein Interactions Akkulak, Hatice; İnce, H. Kerim; Kabasakal, Burak V; Özcan Kabasakal, Süreyya (Orta Doğu Teknik Üniversitesi Enformatik Enstitüsü; 2022-10) Cross-linking mass spectrometry (XL-MS) is an emerging method in structural biology not only for structural elucidation of protein complexes but also for determination of protein-protein interactions. Thus, it is commonly integrated into conventional structural biology techniques, such as Cryo-EM, NMR, and XRay crystallography. XL-MS could be applied to large and dynamic complexes since it is not restricted by sample preparation requirements, and it could capture proteins from their native environment provi...
Using Adaptive Neuro-Fuzzy Inference System for Classification of Microarray Gene Expression Cancer Profiles Haznedar, Bülent; Arslan, Mustafa Turan; Kalınlı, Adem (2018-05-01) Microarray is a technology that enables simultaneously analysis of thousands of genes in DNA structure depending on the advances in biochemistry. With this technology, it has become possible to diagnose and treat heredity diseases by analyzing thousands of gene expression levels. This study proposes an artificial intelligence method, Adaptive neuro-fuzzy inference system (ANFIS), to classify cancer gene expression profiles. The findings obtained with the proposed ANFIS approach are compared with the results...
INTEGRATION OF MACHINE LEARNING AND ENTROPY METHODS FOR POST-GENOME-WIDE ASSOCIATION STUDIES ANALYSIS Yaldız, Burcu; Aydın Son, Yeşim; Department of Medical Informatics (2022-8-31) Non-linear relationships between genotypes play an essential role in understanding the genetic interactions of complex disease traits. Genome-Wide Association Studies (GWAS) have revealed a statistical association between the SNPs in many complex diseases. As GWAS results could not thoroughly explain the genetic background of these disorders, Genome-Wide Interaction Studies started to gain importance. In recent years, various statistical approaches such as entropy-based methods have been suggested for revea...

Citation Formats

A. S. Rifaioğlu, M. J. Martin, R. Atalay, M. V. Atalay, and T. Doğan, “Investigation of Multi-task Deep Neural Networks in Automated Protein Function Prediction,” presented at the ISMB/ECCB 2017: 25th Annual International Conference on Intelligent Systems for Molecular Biology, (21 - 25 Temmuz 2017), Prag, Çek Cumhuriyeti, 2017, Accessed: 00, 2021. [Online]. Available: https://www.iscb.org/ismbeccb2017.