Intrinsic-Extrinsic Convolution and Pooling for Learning on 3D Protein Structures

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F21%3A00118838" target="_blank" >RIV/00216224:14330/21:00118838 - isvavai.cz</a>

Výsledek na webu

<a href="https://openreview.net/forum?id=l0mSUROpwY" target="_blank" >https://openreview.net/forum?id=l0mSUROpwY</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Intrinsic-Extrinsic Convolution and Pooling for Learning on 3D Protein Structures

Popis výsledku v původním jazyce

The result is a paper (16 pages) at International Conference on Learning Representations. Although it is among the very best conferences in CS, since its proceedings do not have an ISBN or ISSN, the result cannot be transferred to the RIV database as a result of type D. The original abstract follows: Proteins perform a large variety of functions in living organisms and thus play a key role in biology. However, commonly used algorithms in protein learning were not specifically designed for protein data, and are therefore not able to capture all relevant structural levels of a protein during learning. To fill this gap, we propose two new learning operators, specifically designed to process protein structures. First, we introduce a novel convolution operator that considers the primary, secondary, and tertiary structure of a protein by using n-D convolutions defined on both the Euclidean distance, as well as multiple geodesic distances between the atoms in a multi-graph. Second, we introduce a set of hierarchical pooling operators that enable multi-scale protein analysis. We further evaluate the accuracy of our algorithms on common downstream tasks, where we outperform state-of-the-art protein learning algorithms.

Název v anglickém jazyce

Intrinsic-Extrinsic Convolution and Pooling for Learning on 3D Protein Structures

Popis výsledku anglicky

The result is a paper (16 pages) at International Conference on Learning Representations. Although it is among the very best conferences in CS, since its proceedings do not have an ISBN or ISSN, the result cannot be transferred to the RIV database as a result of type D. The original abstract follows: Proteins perform a large variety of functions in living organisms and thus play a key role in biology. However, commonly used algorithms in protein learning were not specifically designed for protein data, and are therefore not able to capture all relevant structural levels of a protein during learning. To fill this gap, we propose two new learning operators, specifically designed to process protein structures. First, we introduce a novel convolution operator that considers the primary, secondary, and tertiary structure of a protein by using n-D convolutions defined on both the Euclidean distance, as well as multiple geodesic distances between the atoms in a multi-graph. Second, we introduce a set of hierarchical pooling operators that enable multi-scale protein analysis. We further evaluate the accuracy of our algorithms on common downstream tasks, where we outperform state-of-the-art protein learning algorithms.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

10200 - Computer and information sciences

Projekt

<a href="/cs/project/GC18-18647J" target="_blank" >GC18-18647J: Vizuální analýza interakcí proteinů a ligandů</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů