# DeepGO: Predicting protein functions from sequence and interactions using a deep ontology-aware classifier

## TL;DR

DeepGO employs a deep learning model that integrates protein sequences and interaction networks to predict protein functions within the Gene Ontology framework, significantly outperforming traditional methods like BLAST.

## Contribution

This paper introduces a novel deep ontology-aware classifier that leverages GO structure and multi-modal data for improved protein function prediction.

## Key findings

- Outperforms baseline methods such as BLAST
- Significant improvement in predicting cellular locations
- Effective integration of sequence and interaction data

## Abstract

A large number of protein sequences are becoming available through the application of novel high-throughput sequencing technologies. Experimental functional characterization of these proteins is time-consuming and expensive, and is often only done rigorously for few selected model organisms. Computational function prediction approaches have been suggested to fill this gap. The functions of proteins are classified using the Gene Ontology (GO), which contains over 40,000 classes. Additionally, proteins have multiple functions, making function prediction a large-scale, multi-class, multi-label problem. We have developed a novel method to predict protein function from sequence. We use deep learning to learn features from protein sequences as well as a cross-species protein-protein interaction network. Our approach specifically outputs information in the structure of the GO and utilizes the dependencies between GO classes as background information to construct a deep learning model. We evaluate our method using the standards established by the Computational Assessment of Function Annotation (CAFA) and demonstrate a significant improvement over baseline methods such as BLAST, with significant improvement for predicting cellular locations.